Not in the top 1,000 by GitHub stars
aliyun/qwen-dianjin
Qwen DianJin: LLMs for the Financial Industry by Alibaba Cloud（通义点金：阿里云金融大模型）
History
Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.
View weekly changes
DianJin-SKILLS/claim-expert/claim-adjudication-review/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/claim-expert/claim-adjustment/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/claim-expert/claim-case-registration/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/claim-expert/claim-document-processing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/claim-expert/claim-fraud-detection/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/claim-expert/claim-notification/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/claim-expert/coverage-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/claim-expert/medical-review/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/corporate-banker/credit-due-diligence/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/corporate-banker/credit-industry-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/corporate-banker/equity-penetration-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/corporate-banker/financial-report-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/corporate-banker/pre-visit-plan/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/corporate-banker/product-roadshow/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/corporate-banker/submit-credit-application/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/corporate-banker/visit-memo/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-review-expert/admission-rules-scan/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-review-expert/ai_risk_planning/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-review-expert/credit-case-intake-check/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-review-expert/credit-collateral-risk-mgmt/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-review-expert/credit-related-party-detection/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-review-expert/pre-visit-credit-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-review-expert/reviewer-visit-memo/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-risk-manager/credit-industry-rule-gen/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-risk-manager/credit-large-exposure-mgmt/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-risk-manager/credit-policy-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-risk-manager/credit-risk-cot/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-risk-manager/credit-risk-extraction/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-risk-manager/loan-risk-monitor/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-risk-manager/post-loan-management/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/credit-risk-manager/vlm-verifier/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/financial-engineering-expert/auto-experiment/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/financial-engineering-expert/data-profiling/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/financial-engineering-expert/feature-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/financial-engineering-expert/lr-modeling/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/financial-engineering-expert/lr-tuning/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/financial-engineering-expert/model-comparison/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/financial-engineering-expert/model-explanation/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/financial-engineering-expert/segment-modeling/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/financial-engineering-expert/univariate-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/insurance-agent/activity-planner/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/insurance-agent/application-check/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/insurance-agent/client-consulting/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/insurance-agent/customer-profiling/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/insurance-agent/gap-advisor/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/insurance-agent/health-disclosure-guide/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/insurance-agent/objection-handler/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/insurance-agent/plan-generator/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/insurance-agent/policy-explanation/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/insurance-agent/product-wiki/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/insurance-agent/renewal-assist/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/insurance-agent/social-media/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-advisor/a-market-hotspot-discovery/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-advisor/comparable-company-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-advisor/fund-diagnosis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-advisor/fund-multi-factor-filter/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-advisor/stock-fund-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-advisor/stock-multi-factor-filter/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-advisor/stock-quote-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-advisor/stock-shareholder-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-advisor/stock-tech-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/alternative-factor-mining/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/announcement-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/company-deep-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/company-one-page-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/convertible-bond-valuation/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/earnings-commentary-generator/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/fixed-income-daily-brief/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/fixed-income-overnight-brief/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/fixed-income-rate-review/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/fixed-income-weekly-brief/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/global-finance-brief/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/global-macro-linkage/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/industry-bond-risk-control/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/industry-deep-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/industry-one-page-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/institutional-research-outline/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/lgt-debt-risk-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/macro-asset-allocation/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/macro-daily-briefing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/macro-data-brief/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/macro-policy-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/macro-risk-monitor/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/macro-weekly-briefing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/market-minimal-review/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/market-sentiment-monitor/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/market-trend-style-analysis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/policy-flash-briefing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/quant-daily-brief/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/quant-factor-tracker/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/quant-strategy-quick-backtest/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/quant-weekly-brief/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/sector-allocation/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/strategy-daily-briefing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/investment-researcher/valuation-prosperity-tracking/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/underwriting-expert/conclusion-interpretation/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/underwriting-expert/follow-up-outreach/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/underwriting-expert/health-verification/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/underwriting-expert/medical-assessment/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/underwriting-expert/recording-inspection/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-1_opportunity/client-kyc-profiling/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-1_opportunity/client-opportunity-scan/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-1_opportunity/maturity-client-alert/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-2_outreach/client-care-script/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-2_outreach/hot-product-briefing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-2_outreach/market-event-interpreter/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-2_outreach/market-morning-brief/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-2_outreach/outreach-script-generator/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-2_outreach/wechat-moments-creator/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-3_strategy/client-segmentation-strategy/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-3_strategy/competitor-product-compare/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-3_strategy/sales-sop-navigator/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-4_qa/compliance-risk-qa/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-4_qa/market-insight-qa/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-4_qa/product-knowledge-qa/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-5_diagnosis/fund-deep-research/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-5_diagnosis/portfolio-health-check/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-5_diagnosis/portfolio-risk-radar/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-6_allocation/asset-allocation-optimizer/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-6_allocation/family-financial-planner/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-6_allocation/investment-simulation/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-6_allocation/smart-product-matching/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-7_summary/daily-work-review/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-7_summary/service-summary-generator/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-7_summary/tomorrow-plan-generator/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-8_companion/investor-education-qa/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-8_companion/market-hotspot-digest/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/L2-8_companion/portfolio-alert-narrator/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
DianJin-SKILLS/wealth-copilot/_template/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
e109a7d2da98 → e109a7d2da98
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → e109a7d2da98
No earlier version was observed for comparison.
640GitHub stars
6stars in 7 days
134skills in this repository
1commit in 30 days
Heat
6.2- Star growth, last 7 daysNot available
- Star growth, last 24 hoursNot available
- Rank movement, last 7 daysNot available
- Commits, last 30 daysNot available
- Accounts asking about it through MCP, last 7 daysNot available
- Hot votes this weekNot available
Heat never changes the order by stars, and sponsorship never touches it. How heat works
Stars over time
Individual skills in this repository
- insurance-claim-adjudication-review
DianJin-SKILLS/claim-expert/claim-adjudication-review - insurance-claim-adjustment
DianJin-SKILLS/claim-expert/claim-adjustment - insurance-claim-case-registration
DianJin-SKILLS/claim-expert/claim-case-registration - insurance-claim-document-processing
DianJin-SKILLS/claim-expert/claim-document-processing - insurance-claim-fraud-detection
DianJin-SKILLS/claim-expert/claim-fraud-detection - insurance-claim-notification
DianJin-SKILLS/claim-expert/claim-notification - insurance-claim-coverage-analysis
DianJin-SKILLS/claim-expert/coverage-analysis - insurance-claim-medical-review
DianJin-SKILLS/claim-expert/medical-review - credit-due-diligence
DianJin-SKILLS/corporate-banker/credit-due-diligence - credit-industry-analysis
DianJin-SKILLS/corporate-banker/credit-industry-analysis - equity-penetration-analysis
DianJin-SKILLS/corporate-banker/equity-penetration-analysis - financial-report-analysis
DianJin-SKILLS/corporate-banker/financial-report-analysis - pre-visit-plan
DianJin-SKILLS/corporate-banker/pre-visit-plan - product-roadshow
DianJin-SKILLS/corporate-banker/product-roadshow - submit-credit-application
DianJin-SKILLS/corporate-banker/submit-credit-application - visit-memo
DianJin-SKILLS/corporate-banker/visit-memo - admission-rules-scan
DianJin-SKILLS/credit-review-expert/admission-rules-scan - ai-risk-planning
DianJin-SKILLS/credit-review-expert/ai_risk_planning - credit-case-intake-check
DianJin-SKILLS/credit-review-expert/credit-case-intake-check - credit-collateral-risk-mgmt
DianJin-SKILLS/credit-review-expert/credit-collateral-risk-mgmt - credit-related-party-detection
DianJin-SKILLS/credit-review-expert/credit-related-party-detection - pre-visit-credit-analysis
DianJin-SKILLS/credit-review-expert/pre-visit-credit-analysis - reviewer-visit-memo
DianJin-SKILLS/credit-review-expert/reviewer-visit-memo - credit-industry-rule-gen
DianJin-SKILLS/credit-risk-manager/credit-industry-rule-gen - credit-large-exposure-mgmt
DianJin-SKILLS/credit-risk-manager/credit-large-exposure-mgmt - credit-policy-analysis
DianJin-SKILLS/credit-risk-manager/credit-policy-analysis - credit-risk-cot
DianJin-SKILLS/credit-risk-manager/credit-risk-cot - credit-risk-extraction
DianJin-SKILLS/credit-risk-manager/credit-risk-extraction - loan-risk-monitor
DianJin-SKILLS/credit-risk-manager/loan-risk-monitor - post-loan-management
DianJin-SKILLS/credit-risk-manager/post-loan-management - vlm-verifier
DianJin-SKILLS/credit-risk-manager/vlm-verifier - auto-experiment
DianJin-SKILLS/financial-engineering-expert/auto-experiment - data-profiling
DianJin-SKILLS/financial-engineering-expert/data-profiling - dnn-modeling
DianJin-SKILLS/financial-engineering-expert/dnn-modeling - dnn-tuning
DianJin-SKILLS/financial-engineering-expert/dnn-tuning - feature-analysis
DianJin-SKILLS/financial-engineering-expert/feature-analysis - lr-modeling
DianJin-SKILLS/financial-engineering-expert/lr-modeling - lr-tuning
DianJin-SKILLS/financial-engineering-expert/lr-tuning - model-comparison
DianJin-SKILLS/financial-engineering-expert/model-comparison - model-explanation
DianJin-SKILLS/financial-engineering-expert/model-explanation - segment-modeling
DianJin-SKILLS/financial-engineering-expert/segment-modeling - univariate-analysis
DianJin-SKILLS/financial-engineering-expert/univariate-analysis - xgb-deepmodel
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel - xgb-modeling
DianJin-SKILLS/financial-engineering-expert/xgb-modeling - xgb-tuning
DianJin-SKILLS/financial-engineering-expert/xgb-tuning - insurance-agent-activity-planner
DianJin-SKILLS/insurance-agent/activity-planner - insurance-agent-application-check
DianJin-SKILLS/insurance-agent/application-check - insurance-agent-client-consulting
DianJin-SKILLS/insurance-agent/client-consulting - insurance-agent-customer-profiling
DianJin-SKILLS/insurance-agent/customer-profiling - insurance-agent-gap-advisor
DianJin-SKILLS/insurance-agent/gap-advisor - insurance-agent-health-disclosure-guide
DianJin-SKILLS/insurance-agent/health-disclosure-guide - insurance-agent-objection-handler
DianJin-SKILLS/insurance-agent/objection-handler - insurance-agent-plan-generator
DianJin-SKILLS/insurance-agent/plan-generator - insurance-agent-policy-explanation
DianJin-SKILLS/insurance-agent/policy-explanation - insurance-agent-product-wiki
DianJin-SKILLS/insurance-agent/product-wiki - insurance-agent-renewal-assist
DianJin-SKILLS/insurance-agent/renewal-assist - insurance-agent-social-media
DianJin-SKILLS/insurance-agent/social-media - a-market-hotspot-discovery
DianJin-SKILLS/investment-advisor/a-market-hotspot-discovery - comparable-company-analysis
DianJin-SKILLS/investment-advisor/comparable-company-analysis - fund-diagnosis
DianJin-SKILLS/investment-advisor/fund-diagnosis - fund-multi-factor-filter
DianJin-SKILLS/investment-advisor/fund-multi-factor-filter - stock-fund-analysis
DianJin-SKILLS/investment-advisor/stock-fund-analysis - stock-multi-factor-filter
DianJin-SKILLS/investment-advisor/stock-multi-factor-filter - stock-quote-analysis
DianJin-SKILLS/investment-advisor/stock-quote-analysis - stock-shareholder-analysis
DianJin-SKILLS/investment-advisor/stock-shareholder-analysis - stock-tech-analysis
DianJin-SKILLS/investment-advisor/stock-tech-analysis - alternative-factor-mining
DianJin-SKILLS/investment-researcher/alternative-factor-mining - announcement-analysis
DianJin-SKILLS/investment-researcher/announcement-analysis - company-deep-analysis
DianJin-SKILLS/investment-researcher/company-deep-analysis - company-one-page-analysis
DianJin-SKILLS/investment-researcher/company-one-page-analysis - convertible-bond-valuation
DianJin-SKILLS/investment-researcher/convertible-bond-valuation - earnings-commentary-generator
DianJin-SKILLS/investment-researcher/earnings-commentary-generator - fixed-income-daily-brief
DianJin-SKILLS/investment-researcher/fixed-income-daily-brief - fixed-income-overnight-brief
DianJin-SKILLS/investment-researcher/fixed-income-overnight-brief - fixed-income-rate-review
DianJin-SKILLS/investment-researcher/fixed-income-rate-review - fixed-income-weekly-brief
DianJin-SKILLS/investment-researcher/fixed-income-weekly-brief - global-finance-brief
DianJin-SKILLS/investment-researcher/global-finance-brief - global-macro-linkage
DianJin-SKILLS/investment-researcher/global-macro-linkage - industry-bond-risk-control
DianJin-SKILLS/investment-researcher/industry-bond-risk-control - industry-deep-analysis
DianJin-SKILLS/investment-researcher/industry-deep-analysis - industry-one-page-analysis
DianJin-SKILLS/investment-researcher/industry-one-page-analysis - institutional-research-outline
DianJin-SKILLS/investment-researcher/institutional-research-outline - lgt-debt-risk-analysis
DianJin-SKILLS/investment-researcher/lgt-debt-risk-analysis - macro-asset-allocation
DianJin-SKILLS/investment-researcher/macro-asset-allocation - macro-daily-briefing
DianJin-SKILLS/investment-researcher/macro-daily-briefing - macro-data-brief
DianJin-SKILLS/investment-researcher/macro-data-brief - macro-policy-analysis
DianJin-SKILLS/investment-researcher/macro-policy-analysis - macro-risk-monitor
DianJin-SKILLS/investment-researcher/macro-risk-monitor - macro-weekly-briefing
DianJin-SKILLS/investment-researcher/macro-weekly-briefing - market-minimal-review
DianJin-SKILLS/investment-researcher/market-minimal-review - market-sentiment-monitor
DianJin-SKILLS/investment-researcher/market-sentiment-monitor - market-trend-style-analysis
DianJin-SKILLS/investment-researcher/market-trend-style-analysis - policy-flash-briefing
DianJin-SKILLS/investment-researcher/policy-flash-briefing - quant-daily-brief
DianJin-SKILLS/investment-researcher/quant-daily-brief - quant-factor-tracker
DianJin-SKILLS/investment-researcher/quant-factor-tracker - quant-strategy-quick-backtest
DianJin-SKILLS/investment-researcher/quant-strategy-quick-backtest - quant-weekly-brief
DianJin-SKILLS/investment-researcher/quant-weekly-brief - sector-allocation
DianJin-SKILLS/investment-researcher/sector-allocation - strategy-daily-briefing
DianJin-SKILLS/investment-researcher/strategy-daily-briefing - valuation-prosperity-tracking
DianJin-SKILLS/investment-researcher/valuation-prosperity-tracking - insurance-underwriting-conclusion-interpretation
DianJin-SKILLS/underwriting-expert/conclusion-interpretation - insurance-underwriting-follow-up-outreach
DianJin-SKILLS/underwriting-expert/follow-up-outreach - insurance-underwriting-health-verification
DianJin-SKILLS/underwriting-expert/health-verification - insurance-underwriting-medical-assessment
DianJin-SKILLS/underwriting-expert/medical-assessment - insurance-underwriting-recording-inspection
DianJin-SKILLS/underwriting-expert/recording-inspection - client-kyc-profiling
DianJin-SKILLS/wealth-copilot/L2-1_opportunity/client-kyc-profiling - client-opportunity-scan
DianJin-SKILLS/wealth-copilot/L2-1_opportunity/client-opportunity-scan - maturity-client-alert
DianJin-SKILLS/wealth-copilot/L2-1_opportunity/maturity-client-alert - client-care-script
DianJin-SKILLS/wealth-copilot/L2-2_outreach/client-care-script - hot-product-briefing
DianJin-SKILLS/wealth-copilot/L2-2_outreach/hot-product-briefing - market-event-interpreter
DianJin-SKILLS/wealth-copilot/L2-2_outreach/market-event-interpreter - market-morning-brief
DianJin-SKILLS/wealth-copilot/L2-2_outreach/market-morning-brief - outreach-script-generator
DianJin-SKILLS/wealth-copilot/L2-2_outreach/outreach-script-generator - wechat-moments-creator
DianJin-SKILLS/wealth-copilot/L2-2_outreach/wechat-moments-creator - client-segmentation-strategy
DianJin-SKILLS/wealth-copilot/L2-3_strategy/client-segmentation-strategy - competitor-product-compare
DianJin-SKILLS/wealth-copilot/L2-3_strategy/competitor-product-compare - sales-sop-navigator
DianJin-SKILLS/wealth-copilot/L2-3_strategy/sales-sop-navigator - compliance-risk-qa
DianJin-SKILLS/wealth-copilot/L2-4_qa/compliance-risk-qa - market-insight-qa
DianJin-SKILLS/wealth-copilot/L2-4_qa/market-insight-qa - product-knowledge-qa
DianJin-SKILLS/wealth-copilot/L2-4_qa/product-knowledge-qa - fund-deep-research
DianJin-SKILLS/wealth-copilot/L2-5_diagnosis/fund-deep-research - portfolio-health-check
DianJin-SKILLS/wealth-copilot/L2-5_diagnosis/portfolio-health-check - portfolio-risk-radar
DianJin-SKILLS/wealth-copilot/L2-5_diagnosis/portfolio-risk-radar - asset-allocation-optimizer
DianJin-SKILLS/wealth-copilot/L2-6_allocation/asset-allocation-optimizer - family-financial-planner
DianJin-SKILLS/wealth-copilot/L2-6_allocation/family-financial-planner - investment-simulation
DianJin-SKILLS/wealth-copilot/L2-6_allocation/investment-simulation - smart-product-matching
DianJin-SKILLS/wealth-copilot/L2-6_allocation/smart-product-matching - daily-work-review
DianJin-SKILLS/wealth-copilot/L2-7_summary/daily-work-review - service-summary-generator
DianJin-SKILLS/wealth-copilot/L2-7_summary/service-summary-generator - tomorrow-plan-generator
DianJin-SKILLS/wealth-copilot/L2-7_summary/tomorrow-plan-generator - investor-education-qa
DianJin-SKILLS/wealth-copilot/L2-8_companion/investor-education-qa - market-hotspot-digest
DianJin-SKILLS/wealth-copilot/L2-8_companion/market-hotspot-digest - portfolio-alert-narrator
DianJin-SKILLS/wealth-copilot/L2-8_companion/portfolio-alert-narrator - skill-name-here
DianJin-SKILLS/wealth-copilot/_template
How it works
insurance-claim-adjudication-review
When your agent uses it
理赔核赔全案复审与决策专家。整合全案复审、流程合规检查能力。核查材料齐全性、立案合理性、医审核定准确性、理算计算正确性及作业规范性，综合输出核赔决策（准赔/减赔/拒赔/挂起）。当核赔专员需要对已完成医审和理算的案件进行终审决策、发现疑点需复核时触发。适用于核赔岗全案审核、赔付决策、案件质检场景。触发词：全案复审、怎么判、审核结论汇总。
What it loads into context
Frontmatter
192 tokensAlways available to the agent
SKILL.md
2,564 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 系统依赖
- MCP工具调用
- 输出格式
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset116 bytesSKILL.mdcore8,855 bytesconfig.jsonasset454 bytesrequirements.txtdoc139 bytesscripts/validate_input.pyruns1,789 bytes
Read SKILL.md
# 核赔全案复审与决策
## 角色定义
扮演核赔全案复审专家。你的判断必须以材料齐全性、医审核定准确性、理算计算正确性为依据，绝不猜测缺失信息。
## 触发条件
当满足以下任意条件时触发本技能：
- 核赔专员对已完成医审和理算的案件进行终审
- 案件进入核赔决策节点，需综合全流程审核结果
- 发现医审核定或理算计算存在疑点需复核
- 高金额/高风险案件需要强制核赔复审
insurance-claim-adjustment
When your agent uses it
理赔理算校验与调度专家。对进入理算环节的案件进行准入审查和数据完整性校验，确认理算前置条件满足后，通过MCP调用理算引擎完成实际金额计算，输出标准化理算书。当案件完成医审和责任认定后进入理算环节、需要计算应赔金额、生成理算书时触发。适用于理赔理算前置校验、理算参数校验、理算书生成场景。触发词：赔多少、理算、赔付金额。
What it loads into context
Frontmatter
173 tokensAlways available to the agent
SKILL.md
3,531 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 输出格式
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset520 bytesSKILL.mdcore12,554 bytesconfig.jsonasset995 bytesrequirements.txtdoc258 bytesscripts/validate_input.pyruns1,878 bytes
Read SKILL.md
# 理算校验与调度
## 角色定义
扮演理赔理算校验与调度专家。本技能**不执行实际金额计算**，计算由外部理算引擎完成。职责是：
1. 对理算环节进行**准入审查**（案件状态、前置步骤完成度）
2. 对理算输入数据进行**完整性校验**（费用明细、保单参数、核减结果）
3. 通过 **MCP 调用 `calculation-engine`** 完成实际金额计算
4. 接收理算引擎返回结果，进行**合理性复核**
5. 输出标准化理算书
## 触发条件
当用户提及或需要进行以下场景时触发：insurance-claim-case-registration
When your agent uses it
What it loads into context
Frontmatter
218 tokensAlways available to the agent
SKILL.md
2,839 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 系统依赖
- MCP工具调用
- 输出格式
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset116 bytesSKILL.mdcore10,142 bytesSKILL.mdcore10,142 bytesconfig.jsonasset538 bytesrequirements.txtdoc137 bytesscripts/validate_input.pyruns2,148 bytes
Read SKILL.md
# 理赔报案受理与记录结构化
> 基于报案信息结构化录入、保单状态核验、重复报案检测，生成标准化报案记录并分配案件编号。
## 角色定义
你是一位拥有 10 年经验的理赔报案受理专家。你的登记必须以客户提供的报案信息为依据，绝不猜测缺失信息。
任何报案记录必须附有数据来源标注（客户自述/材料提取/系统查询）。
## 触发条件
- 客户提交理赔报案申请
- 上传报案相关材料或描述事故经过
- 需要生成标准化报案记录或案件编号insurance-claim-document-processing
When your agent uses it
What it loads into context
Frontmatter
261 tokensAlways available to the agent
SKILL.md
2,522 tokensWhen the skill is activated
Reference documents
8,592 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 前置条件
- 指令
- 输出格式
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 补充资源
Scripts executed separately
scripts/analyze_materials.pyscripts/check_completeness.pyscripts/check_consistency.pyscripts/classify_claims.pyscripts/cross_validate.pyscripts/utils.pyscripts/validate_input.py
Files
.env.exampleasset120 bytesSKILL.mdcore9,227 bytesSKILL.mdcore9,227 bytesSKILL.mdcore9,227 bytesSKILL.mdcore9,227 bytesSKILL.mdcore9,227 bytesSKILL.mdcore9,227 bytesSKILL.mdcore9,227 bytesSKILL.mdcore9,227 bytesassets/analysis_output_schema.jsonasset9,642 bytesconfig.jsonasset307 bytesreferences/compliance_rules.mddoc7,124 bytesreferences/reference.mddoc23,919 bytesrequirements.txtdoc161 bytesscripts/analyze_materials.pyruns16,473 bytesscripts/check_completeness.pyruns14,221 bytesscripts/check_consistency.pyruns16,672 bytesscripts/classify_claims.pyruns26,587 bytesscripts/cross_validate.pyruns19,102 bytesscripts/utils.pyruns17,113 bytesscripts/validate_input.pyruns1,953 bytes
Read SKILL.md
# 保险理赔材料智能分析
> 基于前置解析+按需复用架构，对理赔材料进行OCR识别、自动分类、完整性检查、一致性校验、交叉验证和病程时间线梳理。
## 角色定义
你是一位拥有 10 年经验的理赔材料审核专家。你的分析必须以理赔材料为依据，绝不猜测缺失信息。
所有审核结论必须附有数据来源标注和置信度评分。
## 前置条件
在开始工作前，确认以下条件满足：
1. 用户已提供理赔材料图片/PDF文件（含费用清单、病历、发票等）
2. 阿里云百炼 DashScope API Key 已配置insurance-claim-fraud-detection
When your agent uses it
对理赔案件进行多维度欺诈风险检测，综合分析材料一致性、行为模式、历史记录等信息，计算欺诈风险评分，识别虚假理赔、骗保欺诈等可疑信号。当需要评估理赔案件欺诈风险、识别可疑骗保行为、对高风险案件进行预警时触发。适用于理赔欺诈筛查、可疑案件调查决策、反欺诈合规管理场景。触发词：欺诈风险、骗保嫌疑、欺诈评分。
What it loads into context
Frontmatter
169 tokensAlways available to the agent
SKILL.md
4,189 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 旁路监控模式
- 系统依赖
- MCP工具调用
- 输出格式
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset116 bytesSKILL.mdcore14,537 bytesconfig.jsonasset872 bytesrequirements.txtdoc135 bytesscripts/validate_input.pyruns1,739 bytes
Read SKILL.md
# 理赔欺诈风险检测
## 角色定义
扮演反欺诈分析专家。从多个维度对理赔案件进行欺诈风险评估，包括材料一致性检查、行为异常模式识别、关联案件分析等，输出风险评分和可疑信号清单，为调查决策提供依据。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 欺诈
- 骗保
- 风险评估
- 可疑案件insurance-claim-notification
When your agent uses it
理赔结案通知与客户沟通专家。生成标准化的理赔通知书（赔付/拒赔/补件），并配套口头沟通话术与情绪安抚指导，确保书面函件合规、口头沟通专业。当需要出具理赔结果通知、生成拒赔说明书、通知客户补充材料、告知赔付金额，或需要与客户进行拒赔解释、情绪安抚时触发。适用于理赔结案通知生成、拒赔函件起草、补件通知书制作、客户沟通话术辅助场景。触发词：结案通知、通知客户、理赔通知书。
What it loads into context
Frontmatter
197 tokensAlways available to the agent
SKILL.md
4,290 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 上游链路
- 系统依赖
- MCP工具调用
- 输出格式
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset314 bytesSKILL.mdcore15,299 bytesconfig.jsonasset787 bytesrequirements.txtdoc165 bytesscripts/validate_input.pyruns2,191 bytes
Read SKILL.md
# 理赔通知与客户沟通
## 角色定义
扮演理赔结案通知与客户沟通专家。根据理赔审核结论，一步完成两件事：
1. **生成合规通知书** — 赔付通知、拒赔通知、补件通知，符合监管要求，可直接送达客户
2. **配套沟通话术** — 配合通知书的口头沟通话术、情绪安抚指导、协商应对方案
先出正式函件，再配口头话术，确保书面合规、口头专业。
## 触发条件
当用户提及或需要进行以下场景时触发：
insurance-claim-coverage-analysis
When your agent uses it
What it loads into context
Frontmatter
219 tokensAlways available to the agent
SKILL.md
2,990 tokensWhen the skill is activated
Reference documents
537 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 系统依赖
- MCP工具调用
- 输出格式
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
__init__.pylib/__init__.pylib/image_utils.pylib/knowledge.pylib/liability_exclusion.pylib/llm_client.pylib/prompts.pyscripts/validate_input.py
Files
.env.exampleasset264 bytesSKILL.mdcore10,541 bytesTEST.mddoc1,915 bytes__init__.pyruns0 bytesconfig.jsonasset670 byteslib/__init__.pyruns1,083 byteslib/image_utils.pyruns1,548 byteslib/knowledge.pyruns2,002 byteslib/liability_exclusion.pyruns7,790 byteslib/llm_client.pyruns8,406 byteslib/prompts.pyruns2,428 bytesrequirements.txtdoc223 bytesscripts/validate_input.pyruns1,898 bytes
Read SKILL.md
# 责免事项检查
## 角色定义
扮演保险责任免除条款审核专家。通过OCR提取理赔材料内容，与保单免责条款逐条比对，识别可能触发的免责事项，评估拒赔风险等级，输出审核结论。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 免责
- 责免
- 拒赔
- 免责条款insurance-claim-medical-review
When your agent uses it
对理赔费用清单中所有费用项（含检查、手术、耗材、护理、药品、床位费等）进行逐项合理性审核，识别不符合诊断的费用、超标准收费、重复收费；同时审查处方药品与诊断匹配性，识别超适应症用药、剂量异常、配伍禁忌，结合医保目录与临床指南输出核减建议，生成医学审查报告。当用户要求审核理赔费用明细、核定住院费用合理性、识别超标准或重复收费、审核药品合理性时触发。适用于医疗险理赔费用核定、住院费用审查、费用核减决策、理赔用药审核场景。触发词：住院费用、用药合理性、过度收费。
What it loads into context
Frontmatter
233 tokensAlways available to the agent
SKILL.md
4,083 tokensWhen the skill is activated
Reference documents
740 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 系统依赖
- MCP工具调用
- 关联技能
- 输出格式
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
lib/__init__.pylib/drug_indication.pylib/image_utils.pylib/knowledge.pylib/llm_client.pylib/prompts.pyscripts/validate_input.py
Files
.env.exampleasset264 bytesSKILL.mdcore14,315 bytesTEST.mddoc2,656 bytesconfig.jsonasset877 byteslib/__init__.pyruns1,083 byteslib/drug_indication.pyruns7,049 byteslib/image_utils.pyruns1,548 byteslib/knowledge.pyruns2,002 byteslib/llm_client.pyruns8,406 byteslib/prompts.pyruns2,428 bytesrequirements.txtdoc134 bytesscripts/validate_input.pyruns2,046 bytes
Read SKILL.md
# 理赔医学审查
## 角色定义
扮演理赔医审专家。对理赔材料中全部费用项进行逐项合理性审查，覆盖诊疗检查、手术治疗、药品、耗材、护理、床位等所有费用类别，识别与诊断不符、超标准、重复、不必要的费用项；同时对处方药品与患者诊断进行匹配性审查，识别超适应症用药、剂量异常、配伍禁忌等问题，输出结构化审核标记与核减建议，供理算环节直接使用。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 住院费用
- 费用明细
- 用药合理性
- 过度收费credit-due-diligence
When your agent uses it
企业信贷尽职调查报告生成技能。自动获取企业工商、征信、财务、经营等多维度数据，按银行贷前尽调标准进行系统性分析，覆盖企业基本面、公司治理、关联关系、财务健康、经营真实性验证、主体资信、风险评估及授信建议，输出完整结构化尽调报告。触发词包括："尽职调查"、"尽调报告"、"due diligence"、"贷前调查"、"credit investigation"、"做个尽调"、"写尽调报告"。不适用于：贷后管理报告、风险分类调整、不良资产处置、或个人信贷尽调。
What it loads into context
Frontmatter
330 tokensAlways available to the agent
SKILL.md
6,623 tokensWhen the skill is activated
Reference documents
13,165 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 执行流程 (Workflow)
- 合规约束 (Constraints)
- 审计追踪 (Audit Trail)
- 一票否决条件 (Red Lines)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
- 输出格式 (Output Format)
Scripts executed separately
scripts/validate_due_diligence.py
Files
PROVENANCE.mddoc11,017 bytesSKILL.mdcore23,126 bytesassets/due-diligence-report-template.mddoc9,389 bytesprovenance/01-regulatory-review/README.mddoc1,142 bytesprovenance/02-rule-extraction/README.mddoc1,300 bytesprovenance/03-structured-design/README.mddoc1,876 bytesprovenance/04-skill-synthesis/README.mddoc2,132 bytesreferences/industry-analysis-guide.mddoc4,224 bytesreferences/risk-rating-policy.mddoc2,877 bytesscripts/requirements.txtdoc468 bytesscripts/validate_due_diligence.pyruns5,333 bytestests/golden/new-loan-report.mddoc8,149 bytestests/test_cases.yamlasset6,350 bytes
Read SKILL.md
# 企业信贷尽职调查
当用户要求对企业客户进行尽调分析时，按以下步骤系统性执行。
## 目标角色 (Target Role)
- **角色**：对公客户经理、信贷审批官
- **使用场景**：新客户首贷、存量客户续贷、追加授信前的贷前调查阶段
- **输出用途**：系统性评估企业信用风险，为授信审批决策提供依据
- **决策层级**：核心决策支持文件，直接影响授信审批结果
- **执行频率**：每次授信申请前执行一次
## 数据接入 (Data Sources)
credit-industry-analysis
When your agent uses it
尽调报告-信贷行业深度分析技能。对特定行业进行系统性研究，运用PEST宏观环境分析、波特五力竞争模型、产业链价值链拆解等经典框架，从市场规模测算（TAM/SAM/SOM）、竞争格局演变、供需平衡推演、政策环境梳理、技术趋势研判等多维度全面刻画行业现状，并通过情景分析法对行业短中长期发展趋势进行前瞻预判，为授信决策和行业理解提供专业研究支持。触发词包括："行业分析"、"industry analysis"、"行业研究"、"行业评估"、"行业尽调"、"行业准入"、"行业授信政策"、"行业风险排查"。不适用于：企业财务数据分析、股权穿透分析、产品营销方案设计、市场调研、个人信贷行业咨询。
What it loads into context
Frontmatter
399 tokensAlways available to the agent
SKILL.md
6,457 tokensWhen the skill is activated
Reference documents
16,082 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 约束条件 (Constraints)
- 信贷场景专属关注点
- 执行流程 (Workflow)
- 输出格式 (Output Format)
- 审计追踪 (Audit Trail)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/validate_industry.py
Files
PROVENANCE.mddoc10,098 bytesSKILL.mdcore22,337 bytesassets/industry-analysis-template.mddoc11,317 bytesprovenance/01-framework-design/README.mddoc1,326 bytesprovenance/02-expert-interview/README.mddoc1,550 bytesprovenance/03-regulatory-analysis/README.mddoc1,501 bytesprovenance/04-skill-synthesis/README.mddoc2,282 bytesreferences/industry-analysis-guide.mddoc9,777 bytesscripts/requirements.txtdoc22 bytesscripts/validate_industry.pyruns6,426 bytestests/golden/photovoltaic-industry-report.mddoc12,484 bytestests/test_cases.yamlasset5,529 bytes
Read SKILL.md
# 行业分析（Industry Analysis）
## 目标角色 (Target Role)
- **角色**：对公客户经理、信贷审批官、行业研究员、风险经理
- **使用场景**：贷前行业准入评估、授信政策年度调整、特定客户行业背景分析、不良贷款成因分析、新兴行业准入研究
- **输出用途**：生成结构化行业深度分析报告，为授信决策和行业理解提供专业研究支持
- **决策层级**：信贷审批核心参考材料，风险等级 medium，须信贷审批官复核
- **执行频率**：每次授信申请前执行一次，授信政策年度调整时执行一次
## 数据接入 (Data Sources)
### 必需数据
| 数据项 | 来源 | 获取方式 | 敏感级别 |equity-penetration-analysis
When your agent uses it
尽调报告-股权穿透图及关联方分析技能。对目标企业进行股权架构逐层穿透至自然人或终极控制人,绘制Mermaid股权架构图,认定实际控制人及一致行动人,识别法定与隐性关联方,分析关联交易公允性与利益输送风险,输出综合评估报告。触发词包括:"股权穿透"、"equity penetration"、"关联方分析"、"实控人认定"、"关联关系图谱"、"集团架构穿透"、"关联担保网络"、"股权穿透分析"。不适用于:非股权类数据分析、市场调研、行业宏观分析、简单工商信息查询、无具体企业背景的泛泛咨询、个人信贷股权穿透。
What it loads into context
Frontmatter
374 tokensAlways available to the agent
SKILL.md
6,621 tokensWhen the skill is activated
Reference documents
14,282 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 约束条件 (Constraints)
- 分析原则 (Analysis Principles)
- 信贷场景专属关注点
- 执行流程 (Workflow)
- 输出格式 (Output Format)
- 审计追踪 (Audit Trail)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/validate_equity.py
Files
PROVENANCE.mddoc10,360 bytesSKILL.mdcore22,275 bytesassets/equity-penetration-template.mddoc8,755 bytesprovenance/01-methodology-research/README.mddoc1,672 bytesprovenance/02-tool-development/README.mddoc1,371 bytesprovenance/03-regulatory-analysis/README.mddoc1,667 bytesprovenance/04-skill-synthesis/README.mddoc2,175 bytesreferences/equity-penetration-guide.mddoc8,215 bytesscripts/requirements.txtdoc22 bytesscripts/validate_equity.pyruns5,343 bytestests/golden/standard-equity-report.mddoc11,772 bytestests/test_cases.yamlasset4,829 bytes
Read SKILL.md
# 股权穿透图及关联方分析(Equity Penetration Analysis)
## 目标角色 (Target Role)
- **角色**:对公客户经理、信贷审批官、风险经理
- **使用场景**:贷前股权尽调、集团授信穿透、股权质押贷款、IPO/重组前尽调、贷后风险排查
- **输出用途**:生成结构化股权穿透及关联方分析报告,为授信决策和风险评估提供股权与控制权维度的专业支持
- **决策层级**:信贷审批核心参考材料,风险等级 high,需信贷审批官复核
- **执行频率**:每次授信申请前执行一次,贷后风险排查按需执行
## 数据接入 (Data Sources)
### 必需数据
| 数据项 | 来源 | 获取方式 | 敏感级别 |financial-report-analysis
When your agent uses it
尽调报告-财报分析技能。对上市公司或目标企业的财务报表进行全面深度分析。涵盖资产负债表、利润表、现金流量表三大核心报表的解读,通过计算盈利能力、偿债能力、运营效率、成长能力等关键财务指标,结合杜邦分析法拆解ROE驱动力,评估盈利质量(收现比/净现比),进行财务粉饰识别与风险预警,最终输出结构化的财务健康评估报告,为投资决策和经营管理提供数据支撑。当用户上传财务报表、提供公司名称或股票代码并要求进行财务分析时使用此技能。触发词包括:"财报分析"、"财务报表分析"、"financial analysis"、"财务分析"、"分析财报"、"解读财报"、"财报解读"、"财务健康评估"。不适用于:非财务数据分析、市场调研、行业分析、简单数据查询、无具体财务数据的空泛咨询、或个人信贷财务分析。
What it loads into context
Frontmatter
461 tokensAlways available to the agent
SKILL.md
6,956 tokensWhen the skill is activated
Reference documents
19,307 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 执行流程 (Workflow)
- 输出格式 (Output Format)
- 约束条件 (Constraints)
- 分析原则
- 信贷场景专属关注点
- 审计追踪 (Audit Trail)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/validate_financial_report.py
Files
PROVENANCE.mddoc9,951 bytesSKILL.mdcore22,875 bytesassets/financial-report-template.mddoc7,460 bytesprovenance/01-methodology-research/README.mddoc1,599 bytesprovenance/02-tool-development/README.mddoc1,290 bytesprovenance/03-regulatory-analysis/README.mddoc1,642 bytesprovenance/04-skill-synthesis/README.mddoc1,806 bytesreferences/cashflow-pattern-guide.mddoc9,190 bytesreferences/financial-indicators-calculation.mddoc8,376 bytesreferences/fraud-detection-guide.mddoc7,906 bytesscripts/requirements.txtdoc22 bytesscripts/validate_financial_report.pyruns5,046 bytestests/golden/standard-financial-report.mddoc10,045 bytestests/test_cases.yamlasset5,659 bytes
Read SKILL.md
# 财务报表深度分析(Financial Report Analysis)
## 目标角色 (Target Role)
- **角色**:对公客户经理、信贷审批官、风险经理
- **使用场景**:贷前尽调财务分析、年度贷后检视、大额授信审批、重组/展期评估
- **输出用途**:生成结构化财务健康评估报告,为授信决策和投资决策提供数据支撑
- **决策层级**:信贷审批核心参考材料,风险等级 high,需信贷审批官复核
- **执行频率**:每次授信申请前执行一次,贷后每年至少执行一次
## 数据接入 (Data Sources)
### 必需数据
| 数据项 | 来源 | 获取方式 | 敏感级别 |pre-visit-plan
When your agent uses it
客户经理访前规划技能。基于客户全量数据(行内系统、影像档案、NFS文件、风险预警、行业信息),系统性生成访前准备报告,涵盖客户基本面回顾、近期动态分析、拜访目标设定、材料准备清单、沟通策略与风险提示。当用户要求生成拜访前准备报告、制定客户拜访计划、整理客户背景资料、规划营销拜访策略时使用此技能。触发词包括:"拜访前准备"、"访前规划"、"visit planning"、"customer visit preparation"、"拜访计划"、"客户背景分析"。不适用于:拜访后报告生成、授信审批决策、贷款审查、个人信贷访前规划、或无具体客户背景的一般行业研究。
What it loads into context
Frontmatter
331 tokensAlways available to the agent
SKILL.md
5,556 tokensWhen the skill is activated
Reference documents
11,036 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 执行流程 (Workflow)
- 输出格式 (Output Format)
- 合规约束 (Constraints)
- 审计追踪 (Audit Trail)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/validate_visit_plan.py
Files
PROVENANCE.mddoc9,292 bytesSKILL.mdcore19,017 bytesassets/visit-plan-template.mddoc3,008 bytesprovenance/01-requirement-analysis/README.mddoc1,375 bytesprovenance/02-expert-interview/README.mddoc1,849 bytesprovenance/03-regulatory-analysis/README.mddoc2,026 bytesprovenance/04-skill-synthesis/README.mddoc2,827 bytesreferences/industry-classification.mddoc3,241 bytesreferences/visit-strategy-guide.mddoc4,277 bytesscripts/requirements.txtdoc468 bytesscripts/validate_visit_plan.pyruns6,359 bytestests/golden/renew-loan-visit.mddoc4,132 bytestests/golden/risk-inspection-visit.mddoc5,042 bytestests/test_cases.yamlasset4,533 bytes
Read SKILL.md
# 访前规划 (Pre-Visit Planning)
## 目标角色 (Target Role)
- **角色**:对公客户经理
- **使用场景**:拜访客户前的准备阶段,通常在拜访前1-3天执行
- **输出用途**:了解客户全貌,设定拜访目标,准备营销话术和产品推荐方案
- **决策层级**:信息辅助和策略指导,不涉及授信审批决策
- **执行频率**:每次客户拜访前执行一次
---
## 数据接入 (Data Sources)
product-roadshow
When your agent uses it
客户经理产品路演方案生成技能。基于客户经营特征、金融需求、行业背景及竞争格局,生成个性化产品路演方案,涵盖需求诊断、产品组合推荐、竞品对比分析、收益测算与全流程营销话术。当用户需要为特定客户准备产品推介方案、制定营销策略、生成路演材料或设计产品组合时使用此技能。触发词包括:"产品路演"、"product roadshow"、"推介方案"、"营销方案"、"产品推荐"、"roadshow plan"、"产品组合"。不适用于:授信审批决策、贷后管理报告、风险排查、个人信贷产品推介、或无具体客户背景的一般产品咨询。
What it loads into context
Frontmatter
341 tokensAlways available to the agent
SKILL.md
5,091 tokensWhen the skill is activated
Reference documents
9,730 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色(Target Role)
- 数据接入(Data Sources)
- 执行流程(Workflow)
- 合规约束(Constraints)
- 审计追踪(Audit Trail)
- 产品谱系参考
- 输出格式(Output Format)
- 踩坑记录(Gotchas)
- 示例(Examples)
- 非功能范围(Out of Scope)
- 差异化方案策略(根据需求类型自动调整)
- 输入参数说明
Scripts executed separately
scripts/validate_roadshow.py
Files
PROVENANCE.mddoc8,639 bytesSKILL.mdcore17,521 bytesassets/roadshow-plan-template.mddoc5,571 bytesprovenance/01-requirement-analysis/README.mddoc1,608 bytesprovenance/02-expert-interview/README.mddoc2,181 bytesprovenance/03-regulatory-analysis/README.mddoc2,074 bytesprovenance/04-skill-synthesis/README.mddoc2,939 bytesreferences/product-catalog.mddoc8,788 bytesscripts/requirements.txtdoc22 bytesscripts/validate_roadshow.pyruns1,677 bytestests/test_cases.yamlasset5,661 bytes
Read SKILL.md
# 产品路演方案(Product Roadshow)
当用户需要为特定客户生成产品路演方案时,按以下流程执行。
## 目标角色(Target Role)
- **角色**:对公客户经理、产品经理
- **使用场景**:客户拜访前的产品推介准备、专项营销活动筹备、客户需求响应
- **输出用途**:生成个性化产品路演方案,助力客户经理在拜访中实现有效营销
- **决策层级**:信息辅助和营销策略指导,不涉及授信审批决策
- **执行频率**:每次重要客户拜访或专项营销活动前执行一次
---
submit-credit-application
When your agent uses it
授信申请提交辅助技能。自动盘点客户档案中的笔记与材料,对照银行贷前尽调材料标准评估完备性,检查案件冲突与授信到期状态,生成结构化评估报告并引导客户经理确认提交。当用户需要提交授信申请、发起审批流程、续贷申请、追加授信或评估材料是否齐全时使用此技能。触发词包括:"提交授信申请"、"发起审批"、"submit credit"、"续贷申请"、"追加授信"、"评估材料"、"授信申请提交"、"发起授信流程"。不适用于:贷后管理、风险分类调整、不良资产处置、授信审批决策、或无具体客户背景的泛泛咨询。
What it loads into context
Frontmatter
370 tokensAlways available to the agent
SKILL.md
5,190 tokensWhen the skill is activated
Reference documents
9,238 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 约束条件 (Constraints)
- 授信申请材料清单（评估基准）
- 执行流程 (Workflow)
- 输出格式 (Output Format)
- 审计追踪 (Audit Trail)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/validate_credit_application.py
Files
PROVENANCE.mddoc8,989 bytesSKILL.mdcore18,219 bytesSKILL.mdcore18,219 bytesassets/credit-application-template.mddoc4,321 bytesprovenance/01-regulatory-review/README.mddoc1,181 bytesprovenance/02-expert-interview/README.mddoc1,435 bytesprovenance/03-process-design/README.mddoc1,753 bytesprovenance/04-skill-synthesis/README.mddoc2,769 bytesreferences/material-checklist.mddoc5,950 bytesscripts/requirements.txtdoc21 bytesscripts/validate_credit_application.pyruns1,689 bytestests/golden/standard-credit-application.mddoc3,220 bytestests/test_cases.yamlasset6,103 bytes
Read SKILL.md
# 提交授信申请（Submit Credit Application）
当用户要求提交授信申请时，按以下流程执行。
## 目标角色 (Target Role)
- **角色**：对公客户经理、信贷审批官
- **使用场景**：新客户首贷申请、存量客户续贷、追加授信申请
- **输出用途**：评估材料完备性，引导客户经理确认提交，发起授信审批流程
- **决策层级**：授信审批前置环节，直接影响案件是否进入审批流程
- **执行频率**：每次授信申请前执行一次
## 数据接入 (Data Sources)
visit-memo
When your agent uses it
客户经理走访纪要生成技能。将现场拜访的口述观察结构化记录为贷前尽调笔记,支持多轮增量更新与持久化保存。当用户在客户拜访现场口述企业经营状况、厂区观察、财务线索、风险信号、人物印象等信息,或要求记录、整理、更新拜访笔记时使用此技能。触发词包括:"拜访记录"、"走访纪要"、"visit memo"、"现场记录"、"记录拜访"、"整理笔记"、"拜访纪要"、"visit record"。不适用于:拜访前规划、授信审批决策、贷后管理报告、正式尽调报告生成、或无具体客户背景的一般咨询。
What it loads into context
Frontmatter
344 tokensAlways available to the agent
SKILL.md
6,096 tokensWhen the skill is activated
Reference documents
12,245 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 约束条件 (Constraints)
- 审计追踪 (Audit Trail)
- 结构化维度
- 执行流程 (Workflow)
- 交互模式 (Interaction Pattern)
- 输出格式 (Output Format)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 适用范围限制 (Out of Scope)
Scripts executed separately
scripts/validate_visit_memo.py
Files
PROVENANCE.mddoc9,009 bytesSKILL.mdcore20,664 bytesassets/risk-signal-template.mddoc2,820 bytesassets/visit-memo-template.mddoc3,625 bytesprovenance/01-requirement-analysis/README.mddoc1,487 bytesprovenance/02-expert-interview/README.mddoc2,267 bytesprovenance/03-regulatory-analysis/README.mddoc1,333 bytesprovenance/04-skill-synthesis/README.mddoc2,780 bytesreferences/industry-observation-guide.mddoc5,608 bytesreferences/risk-signal-catalog.mddoc5,855 bytesreferences/visit-memo-templates.mddoc3,031 bytesscripts/requirements.txtdoc56 bytesscripts/validate_visit_memo.pyruns4,617 bytestests/golden/standard-visit-memo.mddoc2,156 bytestests/test_cases.yamlasset6,660 bytes
Read SKILL.md
# 拜访记录（Visit Memo）
当用户在客户拜访现场口述信息或要求记录拜访笔记时，按以下流程执行。
## 目标角色 (Target Role)
- **角色**：对公客户经理、风险经理
- **使用场景**：贷前尽调现场拜访、客户实地走访、风险排查拜访
- **输出用途**：将口述观察快速结构化为符合贷前尽调规范的记录
- **决策层级**：贷前调查基础材料，为尽调报告和授信申请提供依据
- **执行频率**：每次拜访现场执行，支持多轮增量更新
## 数据接入 (Data Sources)
admission-rules-scan
When your agent uses it
信贷产品准入规则扫描技能。基于客户画像与目标产品的风控规则库,逐条匹配准入条件,输出"符合/不满足/需确认"三态评估报告,标注一票否决项与差距改进建议。用于授信方案设计阶段的主动预检,帮助客户经理提前识别准入障碍。当用户需要评估客户是否符合某产品准入条件、扫描准入规则、判断客户资质,或说"这个客户能做流贷吗"、"准入扫一下"、"产品准入评估"、"admission scan"、"规则扫描"、"准入检查"时使用此技能。不适用于:贷后管理、风险分类调整、不良资产处置、授信审批决策、或无具体客户背景的一般风控咨询。
What it loads into context
Frontmatter
343 tokensAlways available to the agent
SKILL.md
5,130 tokensWhen the skill is activated
Reference documents
7,634 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 执行流程 (Workflow)
- 输出格式 (Output Format)
- 核心约束 (Constraints)
- 审计追踪 (Audit Trail)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
- 质量要求
- 数据脱敏规则
- 降级策略
- 典型应用场景
- 输入参数说明
Scripts executed separately
scripts/validate_admission_scan.py
Files
PROVENANCE.mddoc8,269 bytesSKILL.mdcore17,223 bytesprovenance/01-regulatory-review/README.mddoc1,762 bytesprovenance/02-rule-extraction/README.mddoc1,961 bytesprovenance/03-structured-design/README.mddoc1,790 bytesprovenance/04-skill-synthesis/README.mddoc2,609 bytesscripts/requirements.txtdoc237 bytesscripts/validate_admission_scan.pyruns5,547 bytestests/golden/TC001_standard_manufacturing.mddoc4,399 bytestests/golden/TC003_A2_rejection.mddoc3,225 bytestests/test_cases.yamlasset5,654 bytes
Read SKILL.md
# L3 准入规则扫描
根据客户画像和目标信贷产品的风控规则库,判断客户是否满足准入条件,输出结构化扫描报告。
---
## 执行流程 (Workflow)
> 交互模式:模式 A - 报告生成型(Report Generation)
当用户要求评估客户产品准入条件时,按以下流程执行:
### 步骤 0:数据确认与验证(先读后写)
ai-risk-planning
When your agent uses it
信贷风控任务规划技能。基于贷款申请信息、输入材料(企业基本信息/尽调报告/年报)及宏观信贷策略,以8大风险筛查维度(固定项)为基础框架,叠加行业风险规则(行业项)和动态专项分析建议(针对项),自顶向下规划风险分析任务和数据采集任务,输出结构化JSON任务规划清单。当用户提供贷款申请信息并要求规划风控审查任务、生成尽调清单、制定信贷审查方案时使用此技能。触发词包括:"风控任务规划"、"risk planning"、"尽调清单生成"、"信贷审查方案"、"风险分析任务"、"风控规划"、"审查任务规划"。不适用于:贷后管理、风险分类调整、不良资产处置、授信审批决策、或无具体贷款申请背景的一般风控咨询。
What it loads into context
Frontmatter
378 tokensAlways available to the agent
SKILL.md
10,013 tokensWhen the skill is activated
Reference documents
9,487 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定位
- 约束条件 (Constraints)
- 规划架构：三层任务体系
- 输入数据说明
- 8大风险筛查维度（固定项框架）
- 行业风险检查项（行业项）
- 动态专项分析建议（针对项）
- 数据采集工具清单
- 审批报告任务
- 执行流程 (Workflow)
- 输出格式 (Output Format)
- 审计追踪 (Audit Trail)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/validate_risk_plan.py
Files
PROVENANCE.mddoc10,129 bytesSKILL.mdcore36,217 bytesprovenance/01-framework-design/README.mddoc1,963 bytesprovenance/02-expert-interview/README.mddoc2,431 bytesprovenance/03-regulatory-analysis/README.mddoc1,533 bytesprovenance/04-skill-synthesis/README.mddoc3,045 bytesscripts/requirements.txtdoc243 bytesscripts/validate_risk_plan.pyruns8,478 bytestests/golden/TC001_standard_manufacturing.mddoc6,093 bytestests/golden/TC002_tech_enterprise.mddoc6,754 bytestests/test_cases.yamlasset7,473 bytes
Read SKILL.md
# 信贷风控任务规划
## 角色定位
专业商业银行信贷风控大脑，具备全行业贷前审查、授信决策、贷后管理经验。熟悉《商业银行法》《贷款通则》《商业银行授信工作尽职指引》《商业银行信用风险内部评级体系监管指引》等监管规定，能够从全局视角识别贷款申请中的实质性风险，自顶向下规划系统化、可执行的风险审查任务体系，服务于贷前尽调、授信审批、信贷政策制定等核心场景。
---
## 约束条件 (Constraints)
> 监管依据:《商业银行法》《贷款通则》《商业银行授信工作尽职指引》《商业银行信用风险内部评级体系监管指引》
1. **自顶向下规划**:遵循“风险识别→分析任务→数据采集”的规划逻辑,先确定风险点,再确定分析方法,最后确定所需数据
2. **动态深度决策**:8大维度必须全部覆盖,但每个维度内的具体检查规则(check_rules)须根据贷款特征动态决定审查深度和侧重点,严禁照抄通用模板credit-case-intake-check
When your agent uses it
授信申请案件进件合规检查技能。对客户提交的授信申请材料进行完整性校验、基本信息有效性核查、申请金额合理性审查,输出材料清单状态(齐全/缺失)、缺失项严重度分级(阻断/建议补充/信息提示),并给出补充指引。触发词包括:"进件检查"、"进件材料检查"、"case intake check"、"帮我检查一下进件材料"、"这个案子材料齐了吗"、"材料校验"、"进件合规检查"、"检查授信材料"。不适用于:贷后管理、风险分类调整、不良资产处置、授信审批决策、客户经理访前分析(请使用pre-visit-credit-analysis)、或无具体案件背景的一般合规咨询。
What it loads into context
Frontmatter
358 tokensAlways available to the agent
SKILL.md
5,422 tokensWhen the skill is activated
Reference documents
9,984 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 执行流程 (Workflow)
- 约束条件 (Constraints)
- 审计追踪 (Audit Trail)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
- 质量要求
- 典型应用场景
- 输入参数说明
Scripts executed separately
scripts/validate_intake_check.py
Files
PROVENANCE.mddoc8,095 bytesSKILL.mdcore18,379 bytesprovenance/01-regulatory-review/README.mddoc1,987 bytesprovenance/02-rule-extraction/README.mddoc2,182 bytesprovenance/03-structured-design/README.mddoc1,563 bytesprovenance/04-skill-synthesis/README.mddoc2,686 bytesscripts/requirements.txtdoc231 bytesscripts/validate_intake_check.pyruns7,179 bytestests/golden/TC001_standard_new_customer.mddoc7,603 bytestests/golden/TC012_I1_rejection.mddoc7,722 bytestests/test_cases.yamlasset8,570 bytes
Read SKILL.md
# 案件进件检查
在授信审查流程启动前,对客户提交的授信申请材料进行完整性校验和基本信息有效性核查,确保案件满足最低进件标准。
---
## 执行流程 (Workflow)
> 交互模式:模式 A - 报告生成型(Report Generation)
当用户要求检查授信申请案件进件材料时,按以下流程执行:
### 步骤 0:数据确认与验证(先读后写)
credit-collateral-risk-mgmt
When your agent uses it
授信押品风险评估与管理技能。覆盖准入筛查、评估合理性审查、登记有效性核验、存续期动态监控、处置合规性检查五大环节,识别押品充足性不足、价值虚高、权属瑕疵、重复担保等风险,输出结构化押品风险报告与分级预警。触发词包括:"押品风险评估"、"押品情况查询"、"抵押率检查"、"collateral risk assessment"、"查一下押品情况"、"这个抵押物有问题吗"、"抵押率超了吗"、"押品合规检查"。不适用于:贷后风险分类调整、不良资产最终核销、押品现场勘估(需第三方评估机构)、或无具体押品背景的一般合规咨询。
What it loads into context
Frontmatter
407 tokensAlways available to the agent
SKILL.md
5,643 tokensWhen the skill is activated
Reference documents
20,804 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 执行流程 (Workflow)
- 约束条件 (Constraints)
- 审计追踪 (Audit Trail)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
- 输出格式 (Output Format)
Scripts executed separately
scripts/validate_collateral.py
Files
PROVENANCE.mddoc8,559 bytesSKILL.mdcore18,735 bytesassets/collateral-risk-report-template.mddoc9,524 bytesprovenance/01-regulatory-review/README.mddoc1,821 bytesprovenance/02-rule-extraction/README.mddoc2,132 bytesprovenance/03-structured-design/README.mddoc2,435 bytesprovenance/04-skill-synthesis/README.mddoc3,200 bytesreferences/collateral-acceptance-criteria.mddoc7,617 bytesreferences/registration-requirements.mddoc8,043 bytesreferences/valuation-standards.mddoc8,056 bytesscripts/requirements.txtdoc225 bytesscripts/validate_collateral.pyruns11,126 bytestests/golden/TC001_standard_pre_investigation.mddoc6,629 bytestests/golden/TC012_C1_rejection.mddoc6,744 bytestests/test_cases.yamlasset11,162 bytes
Read SKILL.md
# 押品全生命周期风险评估与管理
对授信业务涉及的押品(抵押物/质押物),执行准入、评估、登记、监控、处置五环节的风险评估与管理。
## 目标角色 (Target Role)
- **角色**:信贷审批官、风险经理、对公客户经理、贷后管理岗
- **使用场景**:授信审批前押品合规性审查、贷后押品价值监控、风险预警触发后的处置评估
- **输出用途**:生成结构化押品风险报告,作为授信审批决策和贷后风险分类的参考依据
- **决策层级**:高风险决策(押品充足性直接影响授信额度和风险敞口)
- **执行频率**:按需执行(贷前审查/贷后定期监控/风险事件触发)
## 数据接入 (Data Sources)
credit-related-party-detection
When your agent uses it
企业客户关联交易识别与关联风险排查技能。通过股权穿透、管理层任职、家族关系、交易集中度、担保链条等多维度识别关联方,检测资金空转、转移定价、虚构交易、利润转移等异常特征,评估关联风险对授信安全的影响,输出结构化关联方图谱与风险报告。用于贷前关联排查、贷中关联交易监测、贷后风险预警。触发词包括:"关联关系识别"、"关联交易分析"、"关联风险排查"、"related party detection"、"关联图谱"、"查一下这家企业的关联关系"、"关联交易分析一下"、"担保链分析"。不适用于:贷后风险分类调整、不良资产处置、授信审批决策、股权穿透分析(请使用equity-penetration-analysis)、或无具体企业背景的一般合规咨询。
What it loads into context
Frontmatter
449 tokensAlways available to the agent
SKILL.md
6,823 tokensWhen the skill is activated
Reference documents
20,431 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 约束条件 (Constraints)
- 执行步骤
- 一票否决条件（关联交易级）
- 质量要求
- 审计追踪 (Audit Trail)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
- 输出格式 (Output Format)
Scripts executed separately
scripts/validate_related_party.py
Files
PROVENANCE.mddoc9,689 bytesSKILL.mdcore22,825 bytesassets/related-party-report-template.mddoc9,893 bytesprovenance/01-regulatory-review/README.mddoc1,683 bytesprovenance/02-rule-extraction/README.mddoc1,843 bytesprovenance/03-structured-design/README.mddoc2,123 bytesprovenance/04-skill-synthesis/README.mddoc2,988 bytesreferences/related-party-identification-standards.mddoc9,434 bytesreferences/transaction-verification-guide.mddoc11,656 bytesscripts/requirements.txtdoc228 bytesscripts/validate_related_party.pyruns7,985 bytestests/golden/TC001_standard_pre_investigation.mddoc7,153 bytestests/golden/TC012_R1_rejection.mddoc7,667 bytestests/test_cases.yamlasset7,190 bytes
Read SKILL.md
# 关联交易识别与关联风险排查
对企业客户进行全面的关联关系排查和关联交易识别，防范关联风险向授信主体传导。排查须贯穿贷前、贷中、贷后全流程。
---
## 目标角色 (Target Role)
- **角色**: 信贷审批官、风险经理、对公客户经理、合规审查岗
- **使用场景**: 贷前关联排查(新客户首笔授信)、贷中关联交易监测(存续期监控)、贷后风险预警触发时深度排查、大额授信/集团客户授信专项评估
- **输出用途**: 生成结构化关联方图谱与关联风险报告,用于授信审批决策参考、风险定价调整、担保方案设计
- **决策层级**: 高风险(R1-R6触发)须上报合规部门和风险管理部门,中风险须增加担保或压缩额度,低风险纳入常规贷后监测
- **执行频率**: 新客户首笔授信必做、存量客户每半年复核一次、股权变更/高管变动/风险预警触发时临时排查
pre-visit-credit-analysis
When your agent uses it
授信走访前穿透式风险分析技能。整合企业工商、司法、舆情、行内存贷、外部征信五大数据维度,识别红旗/黄旗信号并逐一标注,生成结构化"访前一页纸",辅助客户经理带着问题走访、有的放矢。用于贷前背景调查、存量客户续贷前风险评估、不良预警客户走访准备。触发词包括:"访前分析"、"走访前背景调查"、"访前穿透"、"pre-visit analysis"、"走访前帮我查一下"、"整理一下这个客户的情况"、"摸底企业风险"、"生成访前一页纸"。不适用于:贷后风险分类调整、不良资产处置、授信审批决策、贷后监测(请使用loan-risk-monitor)、反欺诈多模态核验(请使用vlm-verifier)、或无具体客户背景的一般风险咨询。
What it loads into context
Frontmatter
431 tokensAlways available to the agent
SKILL.md
5,532 tokensWhen the skill is activated
Reference documents
14,945 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 约束条件 (Constraints)
- 执行流程 (Workflow)
- 输出格式 (Output Format)
- 审计追踪 (Audit Trail)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/validate_pre_visit.py
Files
PROVENANCE.mddoc8,730 bytesSKILL.mdcore18,514 bytesassets/pre-visit-report-template.mddoc7,436 bytesprovenance/01-framework-design/README.mddoc1,637 bytesprovenance/02-data-integration/README.mddoc1,552 bytesprovenance/03-signal-analysis/README.mddoc2,129 bytesprovenance/04-skill-synthesis/README.mddoc2,767 bytesreferences/pre-visit-data-sources.mddoc6,349 bytesreferences/risk-signal-thresholds.mddoc6,885 bytesscripts/requirements.txtdoc225 bytesscripts/validate_pre_visit.pyruns7,705 bytestests/golden/TC001_standard_new_customer.mddoc4,656 bytestests/golden/TC012_V1_rejection.mddoc4,891 bytestests/test_cases.yamlasset7,061 bytes
Read SKILL.md
# 访前穿透分析
在走访客户之前,对客户进行系统性的背景调查和风险筛查,确保走访目标明确、风险已知。
---
## 目标角色 (Target Role)
- **角色**: 对公客户经理、风险经理、信贷审批官
- **使用场景**: 新客户首次走访前背景调查、存量客户续贷前风险评估、不良预警客户走访准备、大额新增授信走访前摸底
- **输出用途**: 生成结构化"访前一页纸",用于走访计划制定、风险问题清单准备、走访重点方向确定
- **决策层级**: 触发一票否决(V1-V6)须上报风险管理部门,红旗信号须走访时重点核实,黄旗信号纳入常规关注
- **执行频率**: 每次实地走访前必做,新客户首笔授信/存量客户续贷/风险预警触发时强制使用
reviewer-visit-memo
When your agent uses it
授信走访观察纪要生成技能。将走访过程中的口述观察整理为六维度专业走访纪要(现场观感/经营判断/人物印象/资金流向/风险信号/与CM沟通结论),支持与客户经理尽调报告、行内数据三方交叉验证,识别经营场所真实性、经营规模匹配度、资金用途一致性等关键风险。支持多轮增量更新、信息修正与针对性追问,生成可直接归档的第一手审批材料。触发词包括:"走访纪要"、"走访记录"、"reviewer visit memo"、"帮我写个走访纪要"、"刚才走访的情况记一下"、"走访纪要生成"、"审批走访记录"。不适用于:贷后风险分类调整、不良资产处置、授信审批决策、客户经理访前分析(请使用pre-visit-credit-analysis)、或无具体走访背景的一般合规咨询。
What it loads into context
Frontmatter
412 tokensAlways available to the agent
SKILL.md
6,029 tokensWhen the skill is activated
Reference documents
8,571 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 约束条件 (Constraints)
- 执行流程 (Workflow)
- 结构化维度
- 一票否决信号
- 输出格式 (Output Format)
- 交互模式 (Interaction Pattern)
- 上下文管理 (Context Management)
- 审计日志 (Audit Trail)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/validate_visit_memo.py
Files
PROVENANCE.mddoc7,495 bytesSKILL.mdcore19,882 bytesassets/visit-memo-template.mddoc5,476 bytesprovenance/01-knowledge-extraction/README.mddoc1,883 bytesprovenance/02-expert-interview/README.mddoc1,761 bytesprovenance/03-regulatory-analysis/README.mddoc2,022 bytesprovenance/04-skill-synthesis/README.mddoc3,219 bytesreferences/visit-memo-dimensions.mddoc5,470 bytesscripts/requirements.txtdoc118 bytesscripts/validate_visit_memo.pyruns5,343 bytestests/test_cases.yamlasset8,942 bytes
Read SKILL.md
# 走访纪要(审批专员)
当审批专员口述现场走访观察时,按以下规则结构化整理为走访纪要。纪要须客观、具体、可交叉验证,避免主观推断代替客观事实。
---
## 目标角色 (Target Role)
- **角色**: 信贷审批官、审批专员、风险审查岗
- **使用场景**: 新客户首笔授信审批走访、存量客户续贷审批走访、大额新增授信走访、风险预警客户复核走访、CM报告与行内数据偏差较大时的补充走访
- **输出用途**: 生成结构化六维度走访纪要,用于授信审批会决策参考、风险信号识别、CM报告交叉验证
- **决策层级**: 高风险(S1-S6触发)须暂停审批流程并上报,中风险须增加担保或补充核查,低风险纳入常规审批流程
- **执行频率**: 每次实地走访后必做,支持多轮增量更新与信息修正
credit-industry-rule-gen
When your agent uses it
行业信贷风险审查规则生成技能。针对指定行业或产业,结合行业专业知识、产业链结构、政策环境,深度分析并生成结构化的信贷风险审查规则,涵盖基础事实核查类规则和深度推理类规则。触发词包括:"行业信贷规则生成"、"信贷审查项生成"、"行业风控规则"、"风险审批规则"、"行业审查标准"、"补充行业规则"、"industry rule generation"。不适用于:企业个体信用评估、具体授信审批决策、财务数据分析、贷后风险监测。
What it loads into context
Frontmatter
297 tokensAlways available to the agent
SKILL.md
4,693 tokensWhen the skill is activated
Reference documents
6,235 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 执行流程 (Workflow)
- 核心约束 (Constraints)
- 审计追踪 (Audit Trail)
- 输出格式 (Output Format)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/validate_rule_output.py
Files
PROVENANCE.mddoc7,997 bytesSKILL.mdcore16,187 bytesprovenance/01-framework-design/README.mddoc942 bytesprovenance/02-expert-interview/README.mddoc1,075 bytesprovenance/03-regulatory-analysis/README.mddoc1,058 bytesprovenance/04-skill-synthesis/README.mddoc1,589 bytesreferences/fraud-pattern-catalog.mddoc2,852 bytesreferences/industry-lifecycle-guide.mddoc2,051 bytesreferences/regulatory-requirements.mddoc2,557 bytesscripts/requirements.txtdoc133 bytesscripts/validate_rule_output.pyruns15,887 bytestests/test_cases.yamlasset4,441 bytes
Read SKILL.md
# 行业信贷风险审查规则生成
## 目标角色 (Target Role)
- **角色**:资深商业银行信贷风控规则分析师,具备行业研究专家视角
- **使用场景**:针对特定行业或产业进行深度风险解构,从产业链结构、经营特征、监管合规、欺诈模式等多维度切入
- **输出用途**:将行业知识转化为信贷审查人员可直接执行的结构化规则,服务于贷前尽调、授信审批、信贷政策制定
- **决策层级**:风控标准制定,直接影响行业授信准入、审查标准、风险偏好
- **执行频率**:按需执行,通常在新行业授信准入、存量行业规则补充、行业信贷政策修订时执行
## 数据接入 (Data Sources)
### 必需数据
| 数据项 | 来源 | 获取方式 | 敏感级别 |credit-large-exposure-mgmt
When your agent uses it
大额风险暴露集中度监控与压降管理技能。基于银保监会大额风险暴露管理框架,执行全口径风险暴露计量、穿透识别、限额监测、超限预警与压降方案制定,防范集中度风险。触发词包括:"大额风险暴露监控"、"集中度检查"、"风险暴露计量"、"限额监测"、"超限预警"、"压降方案制定"、"查一下集中度"、"这个客户额度超了吗"、"大额风险暴露预警"、"集中度压降"。不适用于:企业个体信用评估、具体授信审批决策、财务数据分析、贷后风险监测。
What it loads into context
Frontmatter
379 tokensAlways available to the agent
SKILL.md
5,272 tokensWhen the skill is activated
Reference documents
9,069 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 执行流程 (Workflow)
- 核心约束 (Constraints)
- 审计追踪 (Audit Trail)
- 输出格式 (Output Format)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/validate_exposure.py
Files
PROVENANCE.mddoc9,706 bytesSKILL.mdcore17,957 bytesassets/disclaimer-template.mddoc1,931 bytesprovenance/01-regulatory-review/README.mddoc893 bytesprovenance/02-rule-extraction/README.mddoc1,189 bytesprovenance/03-structured-design/README.mddoc1,019 bytesprovenance/04-skill-synthesis/README.mddoc1,554 bytesreferences/concentration-analysis-guide.mddoc2,901 bytesreferences/exposure-calculation-guide.mddoc3,745 bytesreferences/group-identification-guide.mddoc2,782 bytesreferences/limit-matrix.mddoc2,426 bytesreferences/mitigation-measures.mddoc2,727 bytesscripts/requirements.txtdoc136 bytesscripts/validate_exposure.pyruns10,257 bytestests/test_cases.yamlasset5,954 bytes
Read SKILL.md
# 大额风险暴露集中度监控与压降管理
对客户授信集中度进行监控和管理,防范大额风险暴露引发的系统性集中度风险。
## 目标角色 (Target Role)
- **角色**:资深商业银行大额风险暴露管理专员
- **使用场景**:单一客户/集团客户/关联方大额风险暴露集中度监控、穿透计量、限额监测、超限预警与压降方案制定
- **输出用途**:为风险管理部、高级管理层、董事会提供集中度风险监控报告,为授信审批提供集中度预检意见
- **决策层级**:风险管控决策,直接影响授信审批、压降方案执行、监管报送
- **执行频率**:每日批量监控+实时新增授信预检,月度/季度/年度定期报告
## 数据接入 (Data Sources)
credit-policy-analysis
When your agent uses it
信贷政策环境分析技能。针对货币政策、监管政策、宏观经济、国际外部环境、区域政策等多维度,搜索最新数据与权威信息,深度分析其对信贷业务的传导路径与影响,输出结构化分析报告及差异化信贷策略建议。触发词包括:"信贷政策分析"、"政策环境分析"、"宏观经济影响分析"、"信贷环境报告"、"货币政策对信贷的影响"、"监管政策解读"、"信贷策略建议"。不适用于:企业财务分析、客户信用评级、具体授信审批决策、贷后风险监测。
What it loads into context
Frontmatter
263 tokensAlways available to the agent
SKILL.md
5,362 tokensWhen the skill is activated
Reference documents
5,470 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 执行流程 (Workflow)
- 分析维度与核心监测指标
- 核心约束
- 审计追踪 (Audit Trail)
- 输出格式 (Output Format)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/validate_policy_analysis.py
Files
PROVENANCE.mddoc7,496 bytesSKILL.mdcore19,111 bytesassets/disclaimer-template.mddoc1,320 bytesprovenance/01-methodology-research/README.mddoc839 bytesprovenance/02-tool-development/README.mddoc1,029 bytesprovenance/03-testing-validation/README.mddoc1,039 bytesprovenance/04-skill-synthesis/README.mddoc1,314 bytesreferences/policy-indicators.mddoc2,822 bytesreferences/risk-alert-rules.mddoc2,864 bytesscripts/requirements.txtdoc127 bytesscripts/validate_policy_analysis.pyruns15,081 bytestests/test_cases.yamlasset4,930 bytes
Read SKILL.md
# 信贷政策环境分析
## 目标角色 (Target Role)
- **角色**：资深商业银行信贷政策环境分析师
- **使用场景**：货币政策、监管政策、宏观经济、国际外部环境、区域政策等多维度分析
- **输出用途**：为授信委员会、行业授信政策制定、客户分层管理提供专业环境分析支撑
- **决策层级**：战略决策支持,直接影响信贷投向、定价策略、风险偏好
- **执行频率**：按需执行,通常在季度/年度政策调整期或重大政策发布后执行
## 数据接入 (Data Sources)
### 必需数据
| 数据项 | 来源 | 获取方式 | 敏感级别 |credit-risk-cot
When your agent uses it
对公授信客户信贷风险逆向思维链(CoT)生成技能。给定原始信贷数据和目标风险点,深度模拟资深风险经理的研判逻辑,构建逻辑严密、证据确凿的推理链,解释为何从该数据能推导出该风险结论。触发词包括:"风险推理链生成"、"思维链分析"、"CoT生成"、"帮我分析一下这个风险点"、"生成思维链"、"这个结论的数据支撑是什么"、"风险点数据支撑分析"。不适用于:无原始数据的纯猜测分析、个人信贷风险评估、非风险类的正向财务分析、贷后风险监测(请使用loan-risk-monitor)。
What it loads into context
Frontmatter
336 tokensAlways available to the agent
SKILL.md
5,701 tokensWhen the skill is activated
Reference documents
8,145 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 术语消歧 (Terminology)
- 执行流程 (Workflow)
- 输出格式 (Output Format)
- 合规红线 (Constraints)
- 审计追踪 (Audit Trail)
- 踩坑记录 (Gotchas)
- 质量要求 (Quality Requirements)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/validate_financial_data.py
Files
PROVENANCE.mddoc10,517 bytesSKILL.mdcore19,837 bytesassets/cot-report-template.mddoc1,386 bytesassets/disclaimer-template.mddoc856 bytesaudit/credit-risk-cot_2026-05-06_refactor_audit.jsonasset3,118 bytesprovenance/01-framework-design/README.mddoc1,736 bytesprovenance/02-methodology-research/README.mddoc1,468 bytesprovenance/03-testing-validation/README.mddoc1,313 bytesprovenance/04-skill-synthesis/README.mddoc1,260 bytesreferences/cot-signal-library.mddoc2,288 bytesreferences/financial-stress-test-params.mddoc1,907 bytesreferences/industry-benchmarks.mddoc1,960 bytesreferences/policy-regulatory-framework.mddoc1,638 bytesscripts/requirements.txtdoc138 bytesscripts/validate_financial_data.pyruns3,314 bytestests/test_cases.yamlasset4,380 bytes
Read SKILL.md
## 目标角色 (Target Role)
- **角色**：信贷风险分析师 / 信审人员
- **使用场景**：贷前尽调、年度贷后复检、风险报告质检、授信审批答辩——需要向审贷委员会提交逻辑完整的风险分析
- **输出用途**：生成结构化推理链，作为风险结论的数据支撑和逻辑证据
- **决策层级**：信息辅助，为授信决策提供逻辑依据，不得直接作为审批结论
- **执行频率**：按需执行，每次授信申请或风险事件触发一次
## 数据接入 (Data Sources)
### 必需数据
| 数据项 | 来源 | 获取方式 | 敏感级别 |
|--------|------|---------|----------|
| 原始信贷数据（财务报表/信贷报告） | 用户上传 | 文件上传 | 内部 |credit-risk-extraction
When your agent uses it
历史报告信贷风险分析框架提取技能。基于企业基本情况描述和已知风险情况，输出企业标签画像与风险结构化分析两大部分，遵循GB/T 4754-2017国标行业分类标准，将抽象风险转化为数据可验证的具体规则。当用户提供企业信贷报告、风险评价文本、授信材料，或要求进行企业风险分析、生成风险提取报告时使用此技能。触发词包括:"风险分析"、"风险提取"、"风险结构化"、"标签画像"、"信贷报告分析"。Do NOT triggering for:贷后风险监测（请使用loan-risk-monitor）、反欺诈多模态核验（请使用vlm-verifier）、要求输出风险分级或审批意见。
What it loads into context
Frontmatter
400 tokensAlways available to the agent
SKILL.md
5,924 tokensWhen the skill is activated
Reference documents
8,511 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 执行流程 (Workflow)
- 输出格式 (Output Format)
- 合规约束 (Constraints)
- 审计追踪 (Audit Trail)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 质量要求
- 典型应用场景
- 输入参数说明
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/validate_risk_extraction.py
Files
PROVENANCE.mddoc9,248 bytesSKILL.mdcore21,362 bytesassets/risk-extraction-report-template.mddoc2,685 bytesprovenance/01-framework-design/README.mddoc1,482 bytesprovenance/02-expert-interview/README.mddoc1,343 bytesprovenance/03-regulatory-analysis/README.mddoc1,358 bytesprovenance/04-skill-synthesis/README.mddoc2,171 bytesreferences/data-sources-priority.mddoc1,484 bytesreferences/scenario-rules.mddoc2,172 bytesreferences/tag-system.mddoc3,452 bytesscripts/requirements.txtdoc134 bytesscripts/validate_risk_extraction.pyruns12,402 bytestests/golden/standard-risk-extraction-report.mddoc5,036 bytestests/test_cases.yamlasset6,182 bytes
Read SKILL.md
# 信贷风险结构化提取
## 目标角色 (Target Role)
- **角色**：风险分析专家 / 信贷审批官 / 风控规则数据化专家
- **场景**：贷前尽调报告结构化、审批委员会材料准备、贷后风险台账维护、风险报告质量审核、行业风险知识库建设
- **输出用途**：生成企业标签画像与风险结构化分析报告，将抽象风险转化为数据可验证的具体规则
- **决策层级**：辅助决策（输出供信审人员参考，不直接用于审批决策）
- **执行频率**：按需执行，通常在贷前尽调或定期风险回顾时调用
## 数据接入 (Data Sources)
### 必需数据
| 数据项 | 来源 | 获取方式 | 敏感级别 |loan-risk-monitor
When your agent uses it
贷后风险监测技能。针对批量借款企业,执行司法、资产、经营、信用及舆情五维度风险信号扫描,自动生成结构化贷后监测报告并提供分级处置建议。触发词包括:"贷后风险监测"、"贷后风险排查"、"定期风险巡检"、"贷后监测报告"、"司法风险搜索"、"资产查封查询"、"经营异常排查"、"债务违约监测"、"负面舆情扫描"。不适用于:企业个体信用评估、具体授信审批决策、财务数据分析、贷前反欺诈审查。
What it loads into context
Frontmatter
334 tokensAlways available to the agent
SKILL.md
3,916 tokensWhen the skill is activated
Reference documents
6,820 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 核心约束 (Constraints)
- 执行流程 (Workflow)
- 审计追踪 (Audit Trail)
- 输出格式 (Output Format)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/validate_monitor_report.py
Files
PROVENANCE.mddoc6,876 bytesSKILL.mdcore13,751 bytesprovenance/01-framework-design/README.mddoc764 bytesprovenance/02-expert-interview/README.mddoc863 bytesprovenance/03-regulatory-analysis/README.mddoc810 bytesprovenance/04-skill-synthesis/README.mddoc1,604 bytesreferences/data-sources-priority.mddoc1,466 bytesreferences/mitigation-measures.mddoc2,336 bytesreferences/monitoring-dimensions.mddoc4,895 bytesreferences/risk-rating-matrix.mddoc1,924 bytesreferences/search-keywords-template.mddoc1,384 bytesscripts/requirements.txtdoc115 bytesscripts/validate_monitor_report.pyruns4,952 bytestests/test_cases.yamlasset5,031 bytes
Read SKILL.md
# 贷后风险监测
## 目标角色 (Target Role)
- **角色**:资深商业银行贷后管理专家
- **使用场景**:批量借款企业贷后风险监测、定期风险巡检、风险信号扫描、分级处置建议生成
- **输出用途**:为贷后管理人员、风控委员会提供准确、及时的风险预警依据和处置建议
- **决策层级**:战术决策支持,直接影响贷后检查频率、风险分类调整、处置策略制定
- **执行频率**:按需执行(风险排查)+定期执行(月度/季度/年度巡检)
## 数据接入 (Data Sources)
### 必需数据
| 数据项 | 来源 | 获取方式 | 敏感级别 |post-loan-management
When your agent uses it
授信客户贷后全周期风险管理技能。执行贷后检查、资金用途核查、财务健康评估、担保有效性审查、预警识别与分类调整、预警处置闭环。触发词包括:"贷后检查"、"贷后管理"、"风险分类调整"、"贷后检查报告"、"做一下贷后检查"、"贷后管理查一下"、"这个客户风险分类需要调吗"、"预警处置方案"、"资金用途核查"。不适用于:贷前尽调、授信审批、个人信贷贷后管理、纯财务分析不涉及贷后场景。
What it loads into context
Frontmatter
336 tokensAlways available to the agent
SKILL.md
5,943 tokensWhen the skill is activated
Reference documents
11,002 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 目标角色 (Target Role)
- 数据接入 (Data Sources)
- 术语消歧 (Terminology)
- 执行流程 (Workflow)
- 输出格式 (Output Format)
- 合规红线 (Constraints)
- 审计追踪 (Audit Trail)
- 踩坑记录 (Gotchas)
- 示例 (Examples)
- 非功能范围 (Out of Scope)
Scripts executed separately
scripts/calculate_financial_ratios.pyscripts/check_fund_usage.pyscripts/evaluate_risk_classification.pyscripts/validate_post_loan_data.py
Files
PROVENANCE.mddoc11,228 bytesSKILL.mdcore20,453 bytesassets/disclaimer-template.mddoc670 bytesassets/post-loan-report-template.mddoc3,318 bytesaudit/post-loan-management_2026-05-06_refactor_audit.jsonasset3,083 bytesprovenance/01-regulatory-review/README.mddoc1,026 bytesprovenance/02-rule-extraction/README.mddoc1,329 bytesprovenance/03-data-model-design/README.mddoc2,216 bytesprovenance/04-skill-synthesis/README.mddoc1,346 bytesreferences/check-frequency-policy.mddoc1,293 bytesreferences/collateral-policy.mddoc1,586 bytesreferences/disposal-escalation-policy.mddoc1,659 bytesreferences/early-warning-indicators.mddoc1,693 bytesreferences/financial-warning-thresholds.mddoc1,119 bytesreferences/five-classification-policy.mddoc1,921 bytesreferences/fund-usage-policy.mddoc2,059 bytesreferences/industry-benchmarks.mddoc1,789 bytesreferences/p1-p6-veto-conditions.mddoc1,494 bytesscripts/calculate_financial_ratios.pyruns4,434 bytesscripts/check_fund_usage.pyruns3,534 bytesscripts/evaluate_risk_classification.pyruns3,533 bytesscripts/requirements.txtdoc68 bytesscripts/validate_post_loan_data.pyruns4,421 bytestests/test_cases.yamlasset4,368 bytes
Read SKILL.md
## 目标角色 (Target Role)
- **角色**：贷后管理岗 / 风险审查人员
- **使用场景**：贷款发放后的持续性贷后管理——首次检查、常规检查、风险预警、分类调整
- **输出用途**：生成结构化贷后检查报告，作为风险分类调整、预警处置、监管检查的依据
- **决策层级**：提供风险分析与分类建议，分类下调为不良类须人工审批
- **执行频率**：按客户风险分类和授信金额矩阵确定（见 references/check-frequency-policy.md）
## 数据接入 (Data Sources)
### 必需数据
| 数据项 | 来源 | 获取方式 | 敏感级别 |
|--------|------|---------|----------|
| 客户基本信息 | 信贷系统 | API: /api/credit/customer | 内部 |vlm-verifier
When your agent uses it
企业信贷跨模态材料核验与分析技能。基于贷款申请材料(图片、文档、流水)与基准数据源,以视觉语言模型(VLM)解析图像证据,结合LLM对文本/结构化数据的分析,实现跨模态数据交叉比对,构建结构化检测点,通过迭代推理验证材料真实性,输出可解释的跨模态核验报告。触发词包括:"材料核验"、"反欺诈审查"、"交叉验证"、"材料真实性核验"、"VLM识别"、"多模态分析"、"欺诈检测"、"跨模态证据核查"。不适用于:个人信贷材料核验、非反欺诈场景(如行业分析/估值建模)、要求输出审批意见或违约概率预测。
What it loads into context
Frontmatter
355 tokensAlways available to the agent
SKILL.md
4,322 tokensWhen the skill is activated
Reference documents
9,620 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Target Role
- Data Sources
- Workflow
- Output Format
- Constraints
- Audit Trail
- Gotchas
- Examples
- Out of Scope
Scripts executed separately
scripts/validate_fraud_report.py
Files
PROVENANCE.mddoc9,689 bytesSKILL.mdcore15,279 bytesassets/verification-report-template.mddoc3,100 bytesprovenance/01-framework-design/README.mddoc1,339 bytesprovenance/02-expert-interview/README.mddoc1,441 bytesprovenance/03-regulatory-analysis/README.mddoc1,422 bytesprovenance/04-skill-synthesis/README.mddoc1,763 bytesreferences/confidence-rules.mddoc1,802 bytesreferences/cross-validation-matrix.mddoc2,653 bytesreferences/data-sources-priority.mddoc1,714 bytesreferences/fraud-patterns.mddoc2,215 bytesscripts/requirements.txtdoc124 bytesscripts/validate_fraud_report.pyruns6,838 bytestests/golden/standard-fraud-verification-report.mddoc5,071 bytestests/test_cases.yamlasset5,643 bytes
Read SKILL.md
# 企业信贷反欺诈多模态交叉验证
## Target Role
- **角色**：信贷审批官 / 反欺诈分析专家 / 风险审查人员
- **使用场景**：贷前申请材料真实性核验、反欺诈交叉验证、多模态证据链构建
- **输出用途**：生成结构化反欺诈评估报告，为信审决策提供客观证据链
- **决策层级**：风险提示信号，需经信贷人员人工复核后使用，不构成审批意见
- **执行频率**：按需执行，每笔企业贷款申请可调用一次
## Data Sources
### 必需数据
| 数据项 | 来源 | 获取方式 | 敏感级别 |auto-experiment
When your agent uses it
【自主实验循环·特征组级探索】当用户说"自动探索哪些特征组有用"、"批量对比不同特征组贡献"、"帮我自主试多种方案"、"自动跑一下"、"试各种特征组合"、"自动特征选择"、"消融分析"时使用。核心能力：自动发现数据中的特征分组，按组执行 独立评估→增量叠加→消融分析→精细筛选 四阶段渐进探索。不做超参数调优，不做单次固定方案训练。与 xgb-tuning 的区别：本 Skill 负责"自主探索特征和方案"，xgb-tuning 负责"调整模型超参数"。与 xgb-modeling 的区别：本 Skill 自主多轮探索，xgb-modeling 按指定方案单次训练。
What it loads into context
Frontmatter
239 tokensAlways available to the agent
SKILL.md
2,099 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 核心理念
- 四阶段探索流程
- 参数说明
- 执行方式
- 常用场景
- 输出内容
- 统计显著性检验
- 与其他技能的关系
- 注意事项
Scripts executed separately
_vendor/__init__.py_vendor/data/__init__.py_vendor/data/loading.py_vendor/data/missing.py_vendor/data/splitting.py_vendor/data/woe_transform.py_vendor/data_utils.py_vendor/evaluation/__init__.py_vendor/evaluation/bootstrap.py_vendor/evaluation/metrics.py_vendor/evaluation/significance.py_vendor/report_artifact.py_vendor/result_emitter.py_vendor/xgb_core/__init__.py_vendor/xgb_core/evaluator.py_vendor/xgb_core/stability.py_vendor/xgb_core/trainer.pyscripts/experiment_loop.pyscripts/hypothesis_generator.pyscripts/run_experiment.pyscripts/session_tracker.pyscripts/significance.py
Files
SKILL.mdcore6,635 bytesSKILL.mdcore6,635 bytesSKILL.mdcore6,635 bytesSKILL.mdcore6,635 bytesSKILL.mdcore6,635 bytes_vendor/__init__.pyruns321 bytes_vendor/data/__init__.pyruns989 bytes_vendor/data/loading.pyruns3,239 bytes_vendor/data/missing.pyruns6,431 bytes_vendor/data/splitting.pyruns20,407 bytes_vendor/data/woe_transform.pyruns8,666 bytes_vendor/data_utils.pyruns1,325 bytes_vendor/evaluation/__init__.pyruns766 bytes_vendor/evaluation/bootstrap.pyruns3,817 bytes_vendor/evaluation/metrics.pyruns4,838 bytes_vendor/evaluation/significance.pyruns5,159 bytes_vendor/report_artifact.pyruns28,991 bytes_vendor/result_emitter.pyruns7,477 bytes_vendor/xgb_core/__init__.pyruns616 bytes_vendor/xgb_core/evaluator.pyruns22,271 bytes_vendor/xgb_core/stability.pyruns9,168 bytes_vendor/xgb_core/trainer.pyruns13,620 bytesscripts/experiment_loop.pyruns24,040 bytesscripts/hypothesis_generator.pyruns17,878 bytesscripts/run_experiment.pyruns19,390 bytesscripts/session_tracker.pyruns7,462 bytesscripts/significance.pyruns452 bytes
Read SKILL.md
# 自主实验循环 (portable)
自动发现数据中的特征分组，执行四阶段渐进式探索实验，每轮完整展示探索逻辑和量化结果。
---
## 核心理念
借鉴 autoresearch 的自主循环思想，升级为**组级探索**：
```
发现特征组 → 组独立评估 → 组间叠加 → 组级消融 → 精细筛选
```
data-profiling
When your agent uses it
【数据轮廓速览】当用户刚上传数据、问"数据有多少行"、"有哪些字段"、"数据长什么样"时使用。描述数据结构（行列数、字段类型、缺失率、分布特征、数据质量），不计算建模指标。无需目标变量。与 feature-analysis 的区别：只做"描述"不做"分析"，不计算IV/PSI/相关性等建模指标。
What it loads into context
Frontmatter
151 tokensAlways available to the agent
SKILL.md
987 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 参数说明
- 执行方式
- 产物示例
- 报告输出结构
- 自适应展示
- 与其他 Skill 的关系
- 注意事项
Scripts executed separately
_vendor/__init__.py_vendor/result_emitter.pyscripts/profiler.py
Files
SKILL.mdcore3,349 bytesSKILL.mdcore3,349 bytes_vendor/__init__.pyruns321 bytes_vendor/result_emitter.pyruns7,477 bytesscripts/profiler.pyruns12,184 bytes
Read SKILL.md
# 数据洞察报告 (portable)
基于 `scripts/profiler.py` 主脚本，对数据集进行轮廓扫描，生成数据概况报告。
---
## 参数说明
| 参数 | 必选 | 默认值 | 说明 |
|------|:----:|--------|------|
| `--data_path` | ✅ | - | 数据文件路径（parquet/csv） |
| `--output` | | `data_profiling_report.md` | 报告输出路径 |
| `--output_dir` | | `./outputs/<ts>` | 产物输出目录 |
| `--output_name` | | `data_profiling_report` | 报告基名（不含扩展名） |dnn-modeling
When your agent uses it
【DNN深度学习建模】当用户需要"深度学习"、"DNN建模"、"神经网络"、"MLP"、"跑个DNN"、"深度模型"时使用。基于 PyTorch 实现多层全连接网络（MLP）进行二分类建模。支持 BatchNorm、Dropout、学习率调度、早停。复用平台已有的三段式数据切分、AUC/KS/BCR评估体系、稳定性分析。前提条件：需要已有数据集。与 xgb-modeling 的区别：本 Skill 使用深度神经网络，适合高维特征交互场景；xgb-modeling 使用 XGBoost 树模型。
What it loads into context
Frontmatter
240 tokensAlways available to the agent
SKILL.md
1,747 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 参数说明
- 执行方式
- 常用场景
- 输出产物
- 与其他建模 Skill 的对比
- 上下游关系
- 注意事项
Scripts executed separately
_vendor/__init__.py_vendor/data/__init__.py_vendor/data/loading.py_vendor/data/missing.py_vendor/data/splitting.py_vendor/data/woe_transform.py_vendor/data_utils.py_vendor/report_artifact.py_vendor/result_emitter.py_vendor/safety_gate.py_vendor/xgb_core/__init__.py_vendor/xgb_core/evaluator.py_vendor/xgb_core/stability.py_vendor/xgb_core/trainer.pyscripts/modeling.py
Files
SKILL.mdcore5,438 bytesSKILL.mdcore5,438 bytesSKILL.mdcore5,438 bytesSKILL.mdcore5,438 bytesSKILL.mdcore5,438 bytesSKILL.mdcore5,438 bytes_vendor/__init__.pyruns321 bytes_vendor/data/__init__.pyruns989 bytes_vendor/data/loading.pyruns3,239 bytes_vendor/data/missing.pyruns6,431 bytes_vendor/data/splitting.pyruns20,407 bytes_vendor/data/woe_transform.pyruns8,666 bytes_vendor/data_utils.pyruns1,325 bytes_vendor/report_artifact.pyruns28,991 bytes_vendor/result_emitter.pyruns7,477 bytes_vendor/safety_gate.pyruns16,308 bytes_vendor/xgb_core/__init__.pyruns616 bytes_vendor/xgb_core/evaluator.pyruns22,271 bytes_vendor/xgb_core/stability.pyruns9,168 bytes_vendor/xgb_core/trainer.pyruns13,620 bytesscripts/modeling.pyruns27,877 bytes
Read SKILL.md
# DNN 深度学习建模 (portable)
基于 PyTorch 实现 MLP (Multi-Layer Perceptron) 进行二分类建模。
**核心流程**：特征标准化 → MLP 训练（BatchNorm + Dropout + Early Stopping） → 概率预测 → 三段式评估
**适用场景**：
- 高维特征交互建模
- 特征间存在复杂非线性关系
- 数据量充足（>10k 样本）
- 对模型性能有极致追求（可与 XGBoost 做 ensemble）
---
dnn-tuning
When your agent uses it
【DNN调参】当用户说"DNN调参"、"深度学习调参"、"调整网络结构"、"调整dropout"、"DNN过拟合"时使用。基于 Optuna TPE 贝叶斯优化，搜索网络架构（层数、宽度、dropout）和训练参数（learning_rate、weight_decay、batch_size），诊断驱动约束搜索空间。前置条件：需先用 dnn-modeling 训练出基线模型。
What it loads into context
Frontmatter
159 tokensAlways available to the agent
SKILL.md
2,082 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 调优流程
- 执行模式
- 参数说明
- 搜索空间
- 诊断驱动策略
- 执行方式
- 调优策略
- 输出格式规范
- 与其他技能的关系
- 注意事项
Scripts executed separately
_vendor/__init__.py_vendor/data/__init__.py_vendor/data/loading.py_vendor/data/missing.py_vendor/data/splitting.py_vendor/data/woe_transform.py_vendor/data_utils.py_vendor/report_artifact.py_vendor/result_emitter.py_vendor/tuning/__init__.py_vendor/tuning/diagnose.py_vendor/tuning/dnn_engine.py_vendor/tuning/engine.py_vendor/tuning/lr_engine.py_vendor/tuning/search_space.py_vendor/tuning/xgb_engine.py_vendor/xgb_core/__init__.py_vendor/xgb_core/evaluator.py_vendor/xgb_core/stability.py_vendor/xgb_core/trainer.pyscripts/tuner.py
Files
SKILL.mdcore6,192 bytesSKILL.mdcore6,192 bytesSKILL.mdcore6,192 bytes_vendor/__init__.pyruns321 bytes_vendor/data/__init__.pyruns989 bytes_vendor/data/loading.pyruns3,239 bytes_vendor/data/missing.pyruns6,431 bytes_vendor/data/splitting.pyruns20,407 bytes_vendor/data/woe_transform.pyruns8,666 bytes_vendor/data_utils.pyruns1,325 bytes_vendor/report_artifact.pyruns28,991 bytes_vendor/result_emitter.pyruns7,477 bytes_vendor/tuning/__init__.pyruns1,165 bytes_vendor/tuning/diagnose.pyruns5,660 bytes_vendor/tuning/dnn_engine.pyruns12,433 bytes_vendor/tuning/engine.pyruns9,339 bytes_vendor/tuning/lr_engine.pyruns7,974 bytes_vendor/tuning/search_space.pyruns6,733 bytes_vendor/tuning/xgb_engine.pyruns4,805 bytes_vendor/xgb_core/__init__.pyruns616 bytes_vendor/xgb_core/evaluator.pyruns22,271 bytes_vendor/xgb_core/stability.pyruns9,168 bytes_vendor/xgb_core/trainer.pyruns13,620 bytesscripts/tuner.pyruns24,267 bytes
Read SKILL.md
# DNN 深度学习参数调优 (portable)
DNN 调参的**唯一入口**，基于 `_vendor/tuning/dnn_engine.DNNTuningEngine`。
核心设计：搜索期间缩减 epochs（加速），诊断驱动动态约束搜索空间。
---
## 调优流程
```
基线 DNN 模型 → 诊断分析(过拟合/欠拟合) → 约束空间构造 → Optuna 搜索(30 epochs) → 最优参数 → 迭代
```
feature-analysis
When your agent uses it
【特征深度分析】当用户需要"全面分析特征"、"生成特征报告"、"看相关性"、"推荐建模方案"、"跑一遍特征"、"特征全景"、"哪些特征有用"时使用。包含四大维度：基础统计、IV值、PSI稳定性、相关性分析，输出四套建模特征方案。前提条件：必须提供目标变量。不做单特征快速探索，不做模型训练。与 univariate-analysis 的区别：本 Skill 是全量特征深度分析+方案推荐，univariate-analysis 聚焦少量特征快速分析。
What it loads into context
Frontmatter
195 tokensAlways available to the agent
SKILL.md
1,439 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 参数说明
- 执行方式
- 常用场景
- 报告输出结构
- 特征筛选方案说明
- 与其他 Skill 的关系
- 注意事项
Scripts executed separately
_vendor/__init__.py_vendor/data/__init__.py_vendor/data/loading.py_vendor/data/missing.py_vendor/data/splitting.py_vendor/data/woe_transform.py_vendor/data_utils.py_vendor/fa_core/__init__.py_vendor/fa_core/correlation.py_vendor/fa_core/iv.py_vendor/fa_core/psi.py_vendor/fa_core/statistics.py_vendor/result_emitter.pyscripts/analyzer.py
Files
SKILL.mdcore4,910 bytesSKILL.mdcore4,910 bytesSKILL.mdcore4,910 bytesSKILL.mdcore4,910 bytesSKILL.mdcore4,910 bytes_vendor/__init__.pyruns321 bytes_vendor/data/__init__.pyruns989 bytes_vendor/data/loading.pyruns3,239 bytes_vendor/data/missing.pyruns6,431 bytes_vendor/data/splitting.pyruns20,407 bytes_vendor/data/woe_transform.pyruns8,666 bytes_vendor/data_utils.pyruns1,325 bytes_vendor/fa_core/__init__.pyruns0 bytes_vendor/fa_core/correlation.pyruns3,701 bytes_vendor/fa_core/iv.pyruns3,889 bytes_vendor/fa_core/psi.pyruns6,542 bytes_vendor/fa_core/statistics.pyruns4,514 bytes_vendor/result_emitter.pyruns7,477 bytesscripts/analyzer.pyruns36,182 bytes
Read SKILL.md
# 建模特征分析报告 (portable)
基于 `scripts/analyzer.py` 主脚本，对数据集特征进行全面分析并生成 Markdown 报告。
---
## 参数说明
| 参数 | 必选 | 默认值 | 说明 |
|------|:----:|--------|------|
| `--data_path` | ✅ | - | 数据文件路径（parquet/csv） |
| `--target` | ✅ | - | 目标变量列名（0/1 二分类） |
| `--exclude_cols` | | - | 排除列，逗号分隔 |
| `--baseline_filter` | | - | PSI 基准数据条件（pandas query） |lr-modeling
When your agent uses it
【LR评分卡建模】当用户需要"逻辑回归"、"LR建模"、"评分卡"、"scorecard"、"WoE建模"、"跑个LR"时使用。基于 WoE 编码 + Logistic Regression 训练标准评分卡模型，输出系数表、分箱WoE映射、评分卡分数转换。复用平台已有的三段式数据切分、AUC/KS/BCR评估体系、稳定性分析。前提条件：需要已有数据集。**不做超参数调优**；不做特征工程。与 xgb-modeling 的区别：本 Skill 使用逻辑回归+WoE编码，输出可解释的评分卡模型；xgb-modeling 使用 XGBoost 树模型。
What it loads into context
Frontmatter
253 tokensAlways available to the agent
SKILL.md
1,816 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 参数说明
- 执行方式
- 常用场景
- 输出产物
- 与其他建模 Skill 的对比
- 上下游关系
- 注意事项
Scripts executed separately
_vendor/__init__.py_vendor/data/__init__.py_vendor/data/loading.py_vendor/data/missing.py_vendor/data/splitting.py_vendor/data/woe_transform.py_vendor/data_utils.py_vendor/lr_cli.py_vendor/report_artifact.py_vendor/result_emitter.py_vendor/safety_gate.py_vendor/xgb_core/__init__.py_vendor/xgb_core/evaluator.py_vendor/xgb_core/stability.py_vendor/xgb_core/trainer.pyscripts/modeling.py
Files
SKILL.mdcore5,663 bytesSKILL.mdcore5,663 bytesSKILL.mdcore5,663 bytesSKILL.mdcore5,663 bytesSKILL.mdcore5,663 bytesSKILL.mdcore5,663 bytes_vendor/__init__.pyruns321 bytes_vendor/data/__init__.pyruns989 bytes_vendor/data/loading.pyruns3,239 bytes_vendor/data/missing.pyruns6,431 bytes_vendor/data/splitting.pyruns20,407 bytes_vendor/data/woe_transform.pyruns8,666 bytes_vendor/data_utils.pyruns1,325 bytes_vendor/lr_cli.pyruns4,287 bytes_vendor/report_artifact.pyruns28,991 bytes_vendor/result_emitter.pyruns7,477 bytes_vendor/safety_gate.pyruns16,308 bytes_vendor/xgb_core/__init__.pyruns616 bytes_vendor/xgb_core/evaluator.pyruns22,271 bytes_vendor/xgb_core/stability.pyruns9,168 bytes_vendor/xgb_core/trainer.pyruns13,620 bytesscripts/modeling.pyruns23,799 bytes
Read SKILL.md
# LR 评分卡建模 (portable)
基于 WoE (Weight of Evidence) 编码 + Logistic Regression 进行二分类评分卡建模。
**核心流程**：原始特征 → WoE 最优分箱编码 → LR 训练 → 评分卡转换
**适用场景**：
- 风控评分卡开发（标准 A/B/C 卡）
- 需要强可解释性的业务场景
- 监管合规要求模型白盒化
---
## 参数说明lr-tuning
When your agent uses it
【LR评分卡调参】当用户说"LR调参"、"评分卡调优"、"调整分箱数"、"调整C值"、"LR过拟合"时使用。基于 Optuna TPE 贝叶斯优化，联合搜索 WoE 分箱参数（max_n_bins、iv_threshold）和 LR 模型参数（C、regularization），诊断驱动约束搜索空间。前置条件：需先用 lr-modeling 训练出基线模型。
What it loads into context
Frontmatter
157 tokensAlways available to the agent
SKILL.md
1,913 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 调优流程
- 执行模式
- 参数说明
- 搜索空间
- 诊断驱动策略
- 执行方式
- 调优策略
- 输出格式规范
- 与其他技能的关系
- 注意事项
Scripts executed separately
_vendor/__init__.py_vendor/data/__init__.py_vendor/data/loading.py_vendor/data/missing.py_vendor/data/splitting.py_vendor/data/woe_transform.py_vendor/data_utils.py_vendor/report_artifact.py_vendor/result_emitter.py_vendor/tuning/__init__.py_vendor/tuning/diagnose.py_vendor/tuning/dnn_engine.py_vendor/tuning/engine.py_vendor/tuning/lr_engine.py_vendor/tuning/search_space.py_vendor/tuning/xgb_engine.py_vendor/xgb_core/__init__.py_vendor/xgb_core/evaluator.py_vendor/xgb_core/stability.py_vendor/xgb_core/trainer.pyscripts/tuner.py
Files
SKILL.mdcore5,685 bytesSKILL.mdcore5,685 bytesSKILL.mdcore5,685 bytes_vendor/__init__.pyruns321 bytes_vendor/data/__init__.pyruns989 bytes_vendor/data/loading.pyruns3,239 bytes_vendor/data/missing.pyruns6,431 bytes_vendor/data/splitting.pyruns20,407 bytes_vendor/data/woe_transform.pyruns8,666 bytes_vendor/data_utils.pyruns1,325 bytes_vendor/report_artifact.pyruns28,991 bytes_vendor/result_emitter.pyruns7,477 bytes_vendor/tuning/__init__.pyruns1,165 bytes_vendor/tuning/diagnose.pyruns5,660 bytes_vendor/tuning/dnn_engine.pyruns12,433 bytes_vendor/tuning/engine.pyruns9,339 bytes_vendor/tuning/lr_engine.pyruns7,974 bytes_vendor/tuning/search_space.pyruns6,733 bytes_vendor/tuning/xgb_engine.pyruns4,805 bytes_vendor/xgb_core/__init__.pyruns616 bytes_vendor/xgb_core/evaluator.pyruns22,271 bytes_vendor/xgb_core/stability.pyruns9,168 bytes_vendor/xgb_core/trainer.pyruns13,620 bytesscripts/tuner.pyruns20,132 bytes
Read SKILL.md
# LR 评分卡参数调优 (portable)
LR 评分卡调参的**唯一入口**，基于 `_vendor/tuning/lr_engine.LRTuningEngine`。
核心设计：WoE 分箱参数与 LR 正则化参数**联合搜索**，确保最优组合。
---
## 调优流程
```
基线 LR 模型 → 诊断分析(过拟合/欠拟合) → 约束空间构造 → Optuna 搜索 → 最优参数 → 迭代
```
model-comparison
When your agent uses it
【多模型效果对比】当用户需要"对比算法"、"多模型对比"、"XGB和LR对比"、"三种算法都试试"、"哪个算法好"、"都跑一下对比看看"、"自动调参后对比"、"公平对比"时使用。基于同一份数据切分，横向运行 XGBoost / LR评分卡 / DNN 等多种算法，补齐 Brier/BCR/真实 PSI 等多维指标，做 DeLong 显著性、样本一致性、特征重叠分析，并产出 **Pareto 前沿候选集 + 客观事实清单**（不做主观加权评分、不做门禁淘汰），最终推荐交由 LLM 基于事实与业务背景推理得出。输出多算法叠加 ROC/KS/Lift 曲线。**公平对比推荐的调用顺序**：xgb-tuning → lr-tuning → dnn-tuning → model-comparison。前提条件：需要已有数据集。
What it loads into context
Frontmatter
288 tokensAlways available to the agent
SKILL.md
2,305 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 推荐工作流：先调参再对比
- 参数说明
- 场景预设（`--scenario`）
- 执行方式
- 对比维度（原始指标，不做加权）
- 统计显著性（DeLong 检验）
- 输出产物
- 与其他技能的关系
- 注意事项
Scripts executed separately
_vendor/__init__.py_vendor/data/__init__.py_vendor/data/loading.py_vendor/data/missing.py_vendor/data/splitting.py_vendor/data/woe_transform.py_vendor/data_utils.py_vendor/evaluation/__init__.py_vendor/evaluation/bootstrap.py_vendor/evaluation/metrics.py_vendor/evaluation/significance.py_vendor/report_artifact.py_vendor/result_emitter.py_vendor/safety_gate.py_vendor/xgb_core/__init__.py_vendor/xgb_core/evaluator.py_vendor/xgb_core/stability.py_vendor/xgb_core/trainer.pyscripts/analysis/__init__.pyscripts/analysis/agreement.pyscripts/analysis/feature_overlap.pyscripts/analysis/metrics.pyscripts/analysis/significance.pyscripts/analysis/stability.pyscripts/comparison.pyscripts/recommend/__init__.pyscripts/recommend/ranker.pyscripts/recommend/scenarios.pyscripts/reporting/__init__.pyscripts/reporting/builder.pyscripts/reporting/renderer.pyscripts/reporting/sections.pyscripts/trainers/__init__.pyscripts/trainers/base.pyscripts/trainers/dnn_trainer.pyscripts/trainers/lr_trainer.pyscripts/trainers/xgb_trainer.pyscripts/tuned_params.py
Files
SKILL.mdcore7,196 bytesSKILL.mdcore7,196 bytesSKILL.mdcore7,196 bytesSKILL.mdcore7,196 bytes_vendor/__init__.pyruns321 bytes_vendor/data/__init__.pyruns989 bytes_vendor/data/loading.pyruns3,239 bytes_vendor/data/missing.pyruns6,431 bytes_vendor/data/splitting.pyruns20,407 bytes_vendor/data/woe_transform.pyruns8,666 bytes_vendor/data_utils.pyruns1,325 bytes_vendor/evaluation/__init__.pyruns766 bytes_vendor/evaluation/bootstrap.pyruns3,817 bytes_vendor/evaluation/metrics.pyruns4,838 bytes_vendor/evaluation/significance.pyruns5,159 bytes_vendor/report_artifact.pyruns28,991 bytes_vendor/result_emitter.pyruns7,477 bytes_vendor/safety_gate.pyruns16,308 bytes_vendor/xgb_core/__init__.pyruns616 bytes_vendor/xgb_core/evaluator.pyruns22,271 bytes_vendor/xgb_core/stability.pyruns9,168 bytes_vendor/xgb_core/trainer.pyruns13,620 bytesscripts/analysis/__init__.pyruns967 bytesscripts/analysis/agreement.pyruns2,220 bytesscripts/analysis/feature_overlap.pyruns2,566 bytesscripts/analysis/metrics.pyruns2,639 bytesscripts/analysis/significance.pyruns3,280 bytesscripts/analysis/stability.pyruns1,621 bytesscripts/comparison.pyruns13,338 bytesscripts/recommend/__init__.pyruns534 bytesscripts/recommend/ranker.pyruns9,447 bytesscripts/recommend/scenarios.pyruns2,128 bytesscripts/reporting/__init__.pyruns401 bytesscripts/reporting/builder.pyruns4,441 bytesscripts/reporting/renderer.pyruns7,107 bytesscripts/reporting/sections.pyruns15,764 bytesscripts/trainers/__init__.pyruns659 bytesscripts/trainers/base.pyruns4,331 bytesscripts/trainers/dnn_trainer.pyruns6,353 bytesscripts/trainers/lr_trainer.pyruns3,884 bytesscripts/trainers/xgb_trainer.pyruns2,503 bytesscripts/tuned_params.pyruns3,859 bytes
Read SKILL.md
# 多模型效果对比 (portable)
基于**同一份数据切分**，横向运行多种算法，统一评估并生成**客观事实报告**。
**设计原则（v3 重要变更）**：
- **只陈述事实，不做主观推荐**：删除了旧版的"五维加权综合评分"与"门禁淘汰"。主观权重本质上将多个不可比指标压成一个数，丢失信息并引入拍脑袋假设；门禁阈值则不同业务差异巨大。
- **Pareto 前沿（无主观权重）**：在 OOT AUC / OOT KS / BCR@10% / KS Gap / PSI 五个方向明确的目标上，识别未被任何算法严格优于的候选集。
- **LLM 推理接手**：报告末尾提供「已知事实清单 + 开放问题」，交由对话中的 AI 结合业务背景推理取舍。
- 同数据同切分 → 消除数据差异对对比结论的干扰
- 多维指标原始陈述 → OOT AUC/KS/BCR/Brier + KS Gap + PSI + 特征数 + 可解释性
- 统计显著性 → DeLong 检验回答"差距是真的还是采样噪声"
---
model-explanation
When your agent uses it
【模型解释】当用户问"为什么模型这样预测"、"哪些特征最重要"、"解释一下这个样本"、"模型怎么做的决策"、"feature importance"、"为什么给这个分数"时使用。基于SHAP的模型解释，支持全局特征重要性、单样本预测解释、特征交互分析。前提条件：需要已训练好的XGBoost模型。不做模型训练，不做特征筛选。与 feature-analysis 的区别：本 Skill 解释"已训练模型的决策逻辑"，feature-analysis 分析"建模前的特征质量"。
What it loads into context
Frontmatter
185 tokensAlways available to the agent
SKILL.md
1,772 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 参数说明
- 执行方式
- 常用场景
- 跨 skill 串联（推荐流程）
- 特征列表来源
- 报告输出结构
- 产物目录
- 与其他技能的关系
- 依赖
- 注意事项
Scripts executed separately
_vendor/__init__.py_vendor/result_emitter.pyscripts/explainer.py../xgb-modeling/scripts/modeling.py
Files
SKILL.mdcore6,027 bytesSKILL.mdcore6,027 bytesSKILL.mdcore6,027 bytesSKILL.mdcore6,027 bytesSKILL.mdcore6,027 bytesSKILL.mdcore6,027 bytesSKILL.mdcore6,027 bytesSKILL.mdcore6,027 bytes_vendor/__init__.pyruns321 bytes_vendor/result_emitter.pyruns7,477 bytesscripts/explainer.pyruns23,003 bytes../xgb-modeling/scripts/modeling.pyruns52,060 bytes
Read SKILL.md
# 模型解释报告 (portable)
基于 SHAP (SHapley Additive exPlanations) 对 XGBoost 模型进行可解释性分析，生成包含可视化图表的 Markdown 报告。
---
## 参数说明
| 参数 | 必选 | 默认值 | 说明 |
|------|:----:|--------|------|
| `--model_path` | ✅ | - | XGBoost 模型文件路径 (.json) |
| `--data_path` | ✅ | - | 数据文件路径（parquet/csv） |
| `--target` | ✅ | - | 目标变量列名 |
| `--features` | | *auto* | 特征列表，逗号分隔。不传时从 `<model>_meta.json` 自动读取 |segment-modeling
When your agent uses it
【分群自主探索】当用户需要"分群建模"、"客群拆分训练"、"探索最优分群方案"、"按xxx分组建模"、"不同客群分别训练"、"分群策略"时使用。支持规则分群（人类先验）、聚类分群（无监督发现）、决策树分群（有监督优化），自动探索最优分群策略并汇总子模型。前提条件：需要已完成基线建模（xgb-modeling）。不做全量样本统一建模，不做超参数调优。与 xgb-modeling 的区别：本 Skill 拆分客群分别建模，xgb-modeling 全量样本统一建模。
What it loads into context
Frontmatter
241 tokensAlways available to the agent
SKILL.md
1,640 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 三种分群策略
- 参数说明
- 执行方式
- 输出内容
- 评估标准
- 与其他技能的关系
- 注意事项
Scripts executed separately
_vendor/__init__.py_vendor/data/__init__.py_vendor/data/loading.py_vendor/data/missing.py_vendor/data/splitting.py_vendor/data/woe_transform.py_vendor/data_utils.py_vendor/result_emitter.py_vendor/xgb_core/__init__.py_vendor/xgb_core/evaluator.py_vendor/xgb_core/stability.py_vendor/xgb_core/trainer.pyscripts/model_merger.pyscripts/run_segment.pyscripts/segment_explorer.pyscripts/segment_hypothesis.pyscripts/segment_strategies.py
Files
SKILL.mdcore5,124 bytesSKILL.mdcore5,124 bytesSKILL.mdcore5,124 bytesSKILL.mdcore5,124 bytesSKILL.mdcore5,124 bytes_vendor/__init__.pyruns321 bytes_vendor/data/__init__.pyruns989 bytes_vendor/data/loading.pyruns3,239 bytes_vendor/data/missing.pyruns6,431 bytes_vendor/data/splitting.pyruns20,407 bytes_vendor/data/woe_transform.pyruns8,666 bytes_vendor/data_utils.pyruns1,325 bytes_vendor/result_emitter.pyruns7,477 bytes_vendor/xgb_core/__init__.pyruns616 bytes_vendor/xgb_core/evaluator.pyruns22,271 bytes_vendor/xgb_core/stability.pyruns9,168 bytes_vendor/xgb_core/trainer.pyruns13,620 bytesscripts/model_merger.pyruns11,672 bytesscripts/run_segment.pyruns20,582 bytesscripts/segment_explorer.pyruns16,538 bytesscripts/segment_hypothesis.pyruns11,158 bytesscripts/segment_strategies.pyruns13,673 bytes
Read SKILL.md
# 分群自主探索 (portable)
在用户指定或 AI 自主发现的分群策略下，拆分客群训练子模型，自动探索最优分群方案。
**核心理念**：
```
探索空间（策略 × 参数 × 组合）
│
▼
Try → Measure → Keep/Discard → Repeat
│
▼
最优分群方案 + 子模型univariate-analysis
When your agent uses it
【单变量分析】当用户提到"分析某个/某些特征"、"看看xxx特征"、"xxx特征分布"、"帮我分箱"、"交叉分布"、"筛选特征"、"分析一下xxx"等涉及少量特征分析的请求时使用。支持两种模式：(1) 无目标变量：等频/等距分箱分布、交叉分布表；(2) 有目标变量：IV值和分箱明细。适用场景关键词：分析特征、看特征、特征分布、分箱、交叉分析、IV值、单个特征、几个特征。与 feature-analysis 的区别：聚焦少量特征的快速分析，不含全量相关性和方案推荐。
What it loads into context
Frontmatter
229 tokensAlways available to the agent
SKILL.md
2,183 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 功能定位
- 参数说明
- 执行方式
- 常用场景
- 报告输出结构
- 分箱说明
- 输出示例
- 与 feature-analysis 的关系
- 注意事项
Scripts executed separately
_vendor/__init__.py_vendor/data/__init__.py_vendor/data/loading.py_vendor/data/missing.py_vendor/data/splitting.py_vendor/data/woe_transform.py_vendor/data_utils.py_vendor/result_emitter.pyscripts/analyzer.pyscripts/binning.pyscripts/cross_distribution.py
Files
SKILL.mdcore6,962 bytesSKILL.mdcore6,962 bytesSKILL.mdcore6,962 bytesSKILL.mdcore6,962 bytesSKILL.mdcore6,962 bytesSKILL.mdcore6,962 bytesSKILL.mdcore6,962 bytesSKILL.mdcore6,962 bytes_vendor/__init__.pyruns321 bytes_vendor/data/__init__.pyruns989 bytes_vendor/data/loading.pyruns3,239 bytes_vendor/data/missing.pyruns6,431 bytes_vendor/data/splitting.pyruns20,407 bytes_vendor/data/woe_transform.pyruns8,666 bytes_vendor/data_utils.pyruns1,325 bytes_vendor/result_emitter.pyruns7,477 bytesscripts/analyzer.pyruns26,989 bytesscripts/binning.pyruns14,152 bytesscripts/cross_distribution.pyruns6,101 bytes
Read SKILL.md
# 单变量分析 (portable)
基于 `scripts/analyzer.py` 主脚本，对指定特征进行单变量级别的分布分析或预测力评估。
---
## 功能定位
| 模式 | 目标变量 | 核心功能 | 使用场景 |
|------|:--------:|----------|----------|
| 数据探索模式 | ❌ 无 | 分布分析、交叉分布 | 刚上传数据、初步了解特征分布 |
| 特征筛选模式 | ✅ 有 | IV值、分箱表、筛选建议 | 建模前快速筛选特征 |
---xgb-deepmodel
When your agent uses it
【DeepModel深度建模】当用户需要"分群集成建模"、"分群+Stacking融合"、"子模型训练+集成"时使用。核心能力：按业务分群条件训练多个 XGBoost 子模型，通过 Stacking 融合成主模型，并与单模型基线对比。适用于"分群建模"、"按客群分别建模"、"子模型融合"、"集成学习"等高级场景。包含三个脚本：sub_trainer（分群子模型训练）、stacker（XGBoost Stacking融合）、comparator（集成vs基线对比）。与 xgb-modeling 的区别：本 Skill 拆分客群分别建模再融合，xgb-modeling 全量样本统一建模。与 segment-modeling 的区别：本 Skill 专注分群+Stacking融合，segment-modeling 专注分群策略探索。
What it loads into context
Frontmatter
257 tokensAlways available to the agent
SKILL.md
2,829 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 阶段1：训练分群子模型
- 阶段2：Stacking 融合
- 阶段3：集成对比报告
- 完整工作流示例
- 质量门控标准
- 与其他技能的关系
- 注意事项
Scripts executed separately
_vendor/__init__.py_vendor/data/__init__.py_vendor/data/loading.py_vendor/data/missing.py_vendor/data/splitting.py_vendor/data/woe_transform.py_vendor/data_utils.py_vendor/report_artifact.py_vendor/result_emitter.py_vendor/safety_gate.py_vendor/tuning/__init__.py_vendor/tuning/diagnose.py_vendor/tuning/dnn_engine.py_vendor/tuning/engine.py_vendor/tuning/lr_engine.py_vendor/tuning/search_space.py_vendor/tuning/xgb_engine.py_vendor/tuning_engine.py_vendor/xgb_cli.py_vendor/xgb_core/__init__.py_vendor/xgb_core/evaluator.py_vendor/xgb_core/stability.py_vendor/xgb_core/trainer.pyscripts/comparator.pyscripts/stacker.pyscripts/sub_trainer.py
Files
SKILL.mdcore9,145 bytesSKILL.mdcore9,145 bytesSKILL.mdcore9,145 bytesSKILL.mdcore9,145 bytesSKILL.mdcore9,145 bytesSKILL.mdcore9,145 bytesSKILL.mdcore9,145 bytesSKILL.mdcore9,145 bytes_vendor/__init__.pyruns321 bytes_vendor/data/__init__.pyruns989 bytes_vendor/data/loading.pyruns3,239 bytes_vendor/data/missing.pyruns6,431 bytes_vendor/data/splitting.pyruns20,407 bytes_vendor/data/woe_transform.pyruns8,666 bytes_vendor/data_utils.pyruns1,325 bytes_vendor/report_artifact.pyruns28,991 bytes_vendor/result_emitter.pyruns7,477 bytes_vendor/safety_gate.pyruns16,308 bytes_vendor/tuning/__init__.pyruns1,165 bytes_vendor/tuning/diagnose.pyruns5,660 bytes_vendor/tuning/dnn_engine.pyruns12,433 bytes_vendor/tuning/engine.pyruns9,339 bytes_vendor/tuning/lr_engine.pyruns7,974 bytes_vendor/tuning/search_space.pyruns6,733 bytes_vendor/tuning/xgb_engine.pyruns4,805 bytes_vendor/tuning_engine.pyruns659 bytes_vendor/xgb_cli.pyruns12,122 bytes_vendor/xgb_core/__init__.pyruns616 bytes_vendor/xgb_core/evaluator.pyruns22,271 bytes_vendor/xgb_core/stability.pyruns9,168 bytes_vendor/xgb_core/trainer.pyruns13,620 bytesscripts/comparator.pyruns21,299 bytesscripts/stacker.pyruns14,571 bytesscripts/sub_trainer.pyruns13,852 bytes
Read SKILL.md
# XGBoost DeepModel — 分群集成建模 (portable)
分群训练多个 XGBoost 子模型，通过 OOF Stacking 融合成主模型，评估集成收益。
---
## 阶段1：训练分群子模型
### 参数说明
> 参数 spec：`_vendor/xgb_cli.py` 中 `domain=deepmodel-sub`。复杂嵌套 JSON（segments / features_per_segment / pos_weight_per_segment）**必须**走 `--config`。
| 参数 | 必选 | 默认值 | 说明 |
|------|:----:|--------|------|xgb-modeling
When your agent uses it
【XGBoost建模】当用户需要"训练模型"、"跑一个基线"、"建个模型试试"、"对比特征方案效果"、"预测一下"、"跑个xgb"、"看AUC/KS/BCR"时使用。支持多特征方案对比、AUC/KS/BCR/Brier评估、Bootstrap置信区间、概率校准曲线、稳定性分析（Mann-Kendall趋势检验）。前提条件：需要已完成特征分析并确定特征方案。**不做超参数调优，调参请用 xgb-tuning**；不做特征工程。与 auto-experiment 的区别：本 Skill 按用户指定方案训练，auto-experiment 自主探索特征组合。
What it loads into context
Frontmatter
256 tokensAlways available to the agent
SKILL.md
2,871 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 参数
- 执行方式
- 产物示例（result.json）
- 跨 skill 串联
- 常用场景
- 报告结构
- 与其他建模 Skill 的对比
- 上下游关系
- 依赖
- 注意事项
Scripts executed separately
_vendor/__init__.py_vendor/data/__init__.py_vendor/data/loading.py_vendor/data/missing.py_vendor/data/splitting.py_vendor/data/woe_transform.py_vendor/data_utils.py_vendor/diagnostic/__init__.py_vendor/diagnostic/fit_check.py_vendor/diagnostic/ks_structure.py_vendor/diagnostic/report.py_vendor/fa_core/__init__.py_vendor/fa_core/correlation.py_vendor/fa_core/iv.py_vendor/fa_core/psi.py_vendor/fa_core/statistics.py_vendor/model_card.py_vendor/report_artifact.py_vendor/result_emitter.py_vendor/safety_gate.py_vendor/tuning/__init__.py_vendor/tuning/diagnose.py_vendor/tuning/search_space.py_vendor/xgb_cli.py_vendor/xgb_core/__init__.py_vendor/xgb_core/evaluator.py_vendor/xgb_core/stability.py_vendor/xgb_core/trainer.pyscripts/feature_selector.pyscripts/modeling.py
Files
SKILL.mdcore9,204 bytesSKILL.mdcore9,204 bytesSKILL.mdcore9,204 bytesSKILL.mdcore9,204 bytesSKILL.mdcore9,204 bytesSKILL.mdcore9,204 bytesSKILL.mdcore9,204 bytesSKILL.mdcore9,204 bytesSKILL.mdcore9,204 bytesSKILL.mdcore9,204 bytes_vendor/__init__.pyruns321 bytes_vendor/data/__init__.pyruns989 bytes_vendor/data/loading.pyruns3,239 bytes_vendor/data/missing.pyruns6,431 bytes_vendor/data/splitting.pyruns20,407 bytes_vendor/data/woe_transform.pyruns8,666 bytes_vendor/data_utils.pyruns1,325 bytes_vendor/diagnostic/__init__.pyruns735 bytes_vendor/diagnostic/fit_check.pyruns3,737 bytes_vendor/diagnostic/ks_structure.pyruns8,163 bytes_vendor/diagnostic/report.pyruns4,656 bytes_vendor/fa_core/__init__.pyruns0 bytes_vendor/fa_core/correlation.pyruns3,701 bytes_vendor/fa_core/iv.pyruns3,889 bytes_vendor/fa_core/psi.pyruns6,542 bytes_vendor/fa_core/statistics.pyruns4,514 bytes_vendor/model_card.pyruns10,232 bytes_vendor/report_artifact.pyruns28,991 bytes_vendor/result_emitter.pyruns7,477 bytes_vendor/safety_gate.pyruns16,308 bytes_vendor/tuning/__init__.pyruns168 bytes_vendor/tuning/diagnose.pyruns5,660 bytes_vendor/tuning/search_space.pyruns6,733 bytes_vendor/xgb_cli.pyruns12,122 bytes_vendor/xgb_core/__init__.pyruns616 bytes_vendor/xgb_core/evaluator.pyruns22,271 bytes_vendor/xgb_core/stability.pyruns9,168 bytes_vendor/xgb_core/trainer.pyruns13,620 bytesscripts/feature_selector.pyruns7,076 bytesscripts/modeling.pyruns52,060 bytes
Read SKILL.md
# XGBoost 建模 (portable)
基于 XGBoost 进行二分类建模，支持自动特征筛选、多方案对比、稳定性分析。可运行在任何 Python 环境，无平台耦合。调参请用 [`xgb-tuning`](./xgb-tuning/SKILL.md)。
**核心流程**：数据加载 → 特征筛选（四大算子）→ 多方案对比 → 最优方案训练 → 三段式评估 → 稳定性分析 → 报告生成
## 参数
> 通用参数 spec 定义在 `_vendor/xgb_cli.py`（domain=`modeling`）。
| 参数 | 必选 | 默认值 | 说明 |
|------|:----:|--------|------|
| `--data_path` / `-d` | ✅ | - | 数据文件路径（parquet/csv） |
| `--target` / `-t` | ✅ | - | 目标变量列名（0/1 二分类） |xgb-tuning
When your agent uses it
【XGBoost超参数调优 — 唯一调参入口】当用户说"帮我调参"、"模型过拟合了怎么办"、"调整learning_rate/max_depth等参数"时使用。核心能力：基于 Optuna TPE 贝叶斯优化 + 诊断驱动的约束搜索（过拟合→收紧树深度上限，欠拟合→抬高树深度下限），每轮输出诊断报告供用户确认。不做特征探索/特征工程，不批量跑多种方案对比。前置条件：需先用 xgb-modeling 训练出基线模型。与 auto-experiment 的区别：本Skill只调整"模型超参数"，auto-experiment 负责"自主探索特征和方案"。
What it loads into context
Frontmatter
224 tokensAlways available to the agent
SKILL.md
5,570 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 调优流程
- 执行模式
- 参数说明
- 基线参数智能推断
- 场景化策略
- 执行方式
- 诊断知识库
- XGBoost 参数语义
- 用户指令理解
- 调优策略
- 输出格式规范
- 5. 调参结论
Scripts executed separately
_vendor/__init__.py_vendor/data/__init__.py_vendor/data/loading.py_vendor/data/missing.py_vendor/data/splitting.py_vendor/data/woe_transform.py_vendor/data_utils.py_vendor/diagnostic/__init__.py_vendor/diagnostic/fit_check.py_vendor/diagnostic/ks_structure.py_vendor/diagnostic/report.py_vendor/model_card.py_vendor/report_artifact.py_vendor/result_emitter.py_vendor/safety_gate.py_vendor/tuning/__init__.py_vendor/tuning/diagnose.py_vendor/tuning/dnn_engine.py_vendor/tuning/engine.py_vendor/tuning/lr_engine.py_vendor/tuning/search_space.py_vendor/tuning/xgb_engine.py_vendor/tuning_engine.py_vendor/xgb_cli.py_vendor/xgb_core/__init__.py_vendor/xgb_core/evaluator.py_vendor/xgb_core/stability.py_vendor/xgb_core/trainer.pyscripts/tuner.py
Files
SKILL.mdcore16,803 bytesSKILL.mdcore16,803 bytesSKILL.mdcore16,803 bytesSKILL.mdcore16,803 bytes_vendor/__init__.pyruns321 bytes_vendor/data/__init__.pyruns989 bytes_vendor/data/loading.pyruns3,239 bytes_vendor/data/missing.pyruns6,431 bytes_vendor/data/splitting.pyruns20,407 bytes_vendor/data/woe_transform.pyruns8,666 bytes_vendor/data_utils.pyruns1,325 bytes_vendor/diagnostic/__init__.pyruns735 bytes_vendor/diagnostic/fit_check.pyruns3,737 bytes_vendor/diagnostic/ks_structure.pyruns8,163 bytes_vendor/diagnostic/report.pyruns4,656 bytes_vendor/model_card.pyruns10,232 bytes_vendor/report_artifact.pyruns28,991 bytes_vendor/result_emitter.pyruns7,477 bytes_vendor/safety_gate.pyruns16,308 bytes_vendor/tuning/__init__.pyruns1,165 bytes_vendor/tuning/diagnose.pyruns5,660 bytes_vendor/tuning/dnn_engine.pyruns12,433 bytes_vendor/tuning/engine.pyruns9,339 bytes_vendor/tuning/lr_engine.pyruns7,974 bytes_vendor/tuning/search_space.pyruns6,733 bytes_vendor/tuning/xgb_engine.pyruns4,805 bytes_vendor/tuning_engine.pyruns659 bytes_vendor/xgb_cli.pyruns12,122 bytes_vendor/xgb_core/__init__.pyruns616 bytes_vendor/xgb_core/evaluator.pyruns22,271 bytes_vendor/xgb_core/stability.pyruns9,168 bytes_vendor/xgb_core/trainer.pyruns13,620 bytesscripts/tuner.pyruns41,598 bytes
Read SKILL.md
# XGBoost 参数调优 (portable)
XGBoost 调参的**唯一入口**，基于 `_vendor/tuning_engine.TuningEngine`。核心设计：
1. **基线参数智能推断** — 根据数据特征推荐合理起点
2. **模型状态诊断** — 过拟合/欠拟合判定（`diagnose_model`）
3. **约束式贝叶斯搜索** — 诊断结论定向收缩 Optuna 搜索空间
4. **用户知识融合** — 接受用户领域经验调整策略
---
## 调优流程
```insurance-agent-activity-planner
When your agent uses it
为保险代理人提供每日/每周/每月展业活动计划制定的专业技能，帮助代理人科学管理拜访量、跟进节点和业绩目标。当代理人需要制定展业计划、管理客户拜访节奏、追踪业绩达成进度、分解月度目标时触发。适用于月初目标分解、周计划制定、每日行程安排、业绩落后时的追赶计划等场景。 触发词：今日复盘、明日计划、拜访安排。
What it loads into context
Frontmatter
163 tokensAlways available to the agent
SKILL.md
3,805 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 输出格式
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset116 bytesSKILL.mdcore12,922 bytesconfig.jsonasset420 bytesrequirements.txtdoc136 bytesscripts/validate_input.pyruns1,863 bytes
Read SKILL.md
# 展业活动计划制定
## 角色定义
扮演保险代理人展业规划专家。
## 触发条件
当代理人需要：
- 制定月度/周度/每日展业计划
- 将业绩目标分解为具体行动
- 管理客户拜访频次与跟进节点
- 评估当前活动量是否足够支撑目标
- 复盘当日展业成果并制定次日拜访计划
- 查看待跟进客户的关键节点提醒insurance-agent-application-check
When your agent uses it
为保险代理人提供投保前资料完整性核查与投保单填写辅助的专业技能，帮助代理人在提交投保前自查，避免因材料缺失或填写错误导致退单。当代理人在帮客户投保时需要核查材料是否齐全、投保单填写是否正确、健康告知是否遗漏时触发。适用于投保前自查、标准化交件流程、新人代理人规范化操作指导等场景。 触发词：材料检查、材料齐不齐、投保单核查。
What it loads into context
Frontmatter
164 tokensAlways available to the agent
SKILL.md
2,657 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 转交核保域数据契约
- 输出格式
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset116 bytesSKILL.mdcore9,306 bytesconfig.jsonasset377 bytesrequirements.txtdoc137 bytesscripts/validate_input.pyruns1,946 bytes
Read SKILL.md
# 投保前核查
## 角色定义
扮演保险投保前核查专家。
## 触发条件
当代理人需要：
- 提交投保前核查材料是否齐全
- 检查投保单各字段是否填写完整正确
- 确认健康告知填写逻辑是否符合要求
- 避免退单影响客户体验
## 前置条件insurance-agent-client-consulting
When your agent uses it
为保险代理人提供客户咨询与需求分析的专业指导。当进行客户初次沟通、需求挖掘、家庭情况了解、风险偏好评估时使用。帮助代理人以专业、温暖的方式建立信任，准确识别客户真实需求。 触发词：客户咨询、挖掘需求、客户想了解。
What it loads into context
Frontmatter
123 tokensAlways available to the agent
SKILL.md
3,350 tokensWhen the skill is activated
Reference documents
602 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 输出格式
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset116 bytesSKILL.mdcore11,828 bytesTEST.mddoc2,157 bytesconfig.jsonasset744 bytesrequirements.txtdoc138 bytesscripts/validate_input.pyruns1,587 bytes
Read SKILL.md
# Agent客户咨询与需求分析
## 角色定义
扮演保险代理人客户咨询专家。以专业、温暖的方式与客户建立信任，通过系统化流程准确识别客户真实保障需求与财务目标。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 客户咨询
- 需求分析
- 初次沟通
- 风险评估
insurance-agent-customer-profiling
When your agent uses it
保险代理人智能客户画像360度分析工具。根据客户保单数据、理赔记录、交互行为等信息，自动生成动态标签体系、保障缺口矩阵、生命周期定位和展业策略推荐。当用户提到客户画像、客户分析、客户标签、保障缺口、展业策略、客户360、客户洞察、保单分析时使用。 触发词：客户画像、客户情况、画像更新。
What it loads into context
Frontmatter
141 tokensAlways available to the agent
SKILL.md
3,343 tokensWhen the skill is activated
Reference documents
3,912 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 输出格式
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/analyze_profile.pyscripts/validate_input.py
Files
.env.exampleasset116 bytesSKILL.mdcore12,324 bytesconfig.jsonasset736 bytesdemo-data/demo_customers.jsonasset30,181 bytesexamples.mddoc3,061 bytesrequirements.txtdoc138 bytesscripts/analyze_profile.pyruns29,047 bytesscripts/validate_input.pyruns1,870 bytestemplates/profile_report.mddoc3,209 bytestemplates/strategy_template.mddoc2,789 bytestemplates/tagging_rules.mddoc3,041 bytes
Read SKILL.md
# 智能客户画像分析
## 角色定义
扮演保险智能客户画像分析专家。基于客户保单数据、理赔记录、交互行为等信息，自动生成六维动态标签体系、保障缺口矩阵、生命周期定位和个性化展业策略推荐。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 客户画像
- 客户分析
- 客户标签
- 客户360
insurance-agent-gap-advisor
When your agent uses it
保险保障缺口分析与配置建议专家。基于用户画像、家庭责任、财务状况、健康信息及现有保单持仓，识别医疗、重疾、寿险、意外等保障缺口，判断配置重复与错配，生成带有完整计算过程和推理依据的可解释性保险配置建议报告（DOCX/PDF）。当用户提及保障缺口、保险分析、保单诊断、保险配置建议、保障规划时触发。 触发词：保障缺口、加什么保、缺口扫描。
What it loads into context
Frontmatter
169 tokensAlways available to the agent
SKILL.md
2,095 tokensWhen the skill is activated
Reference documents
7,888 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 核心理念
- 触发条件
- 前置条件
- 工作流程
- 输出格式
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 参考资料
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset116 bytesSKILL.mdcore7,970 bytesTEST.mddoc2,194 bytesassets/workflow.mddoc8,878 bytesconfig.jsonasset421 bytesreference.mddoc8,865 bytesreport-template.mddoc5,244 bytesrequirements.txtdoc132 bytesscripts/validate_input.pyruns1,897 bytes
Read SKILL.md
# 保险保障缺口分析与配置建议
## 角色定义
扮演保险保障缺口分析与配置建议专家。基于客户画像、家庭责任和财务状况，通过可解释性推理识别保障缺口，生成透明、可信的保险配置建议报告。
## 核心理念
本技能的核心差异是**可解释性推理**：每一条缺口判断和配置建议都必须附带透明的计算过程、数据来源和推理链条，让用户清楚地理解"为什么需要这个保障"以及"保额是怎么算出来的"。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 保障缺口insurance-agent-health-disclosure-guide
When your agent uses it
为保险代理人提供健康告知填写指导与疑难解析的专业技能，帮助代理人引导客户如实、正确地完成健康告知，避免隐瞒重要信息导致日后理赔纠纷。当客户有既往病史或健康状况需要告知、健康告知问卷有疑难问题不知如何填写、担心健康告知影响核保结果时触发。适用于有既往病史的客户投保前健康告知填写、健康告知疑难问题解答、如实告知原则宣导等场景。 触发词：健康告知、结节怎么告知、告知引导。
What it loads into context
Frontmatter
192 tokensAlways available to the agent
SKILL.md
2,705 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 输出格式
- 系统依赖
- MCP工具调用
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 关联技能
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset116 bytesSKILL.mdcore9,346 bytesconfig.jsonasset447 bytesrequirements.txtdoc143 bytesscripts/validate_input.pyruns1,411 bytes
Read SKILL.md
# 健康告知填写指导
## 角色定义
扮演保险投保顾问专家。为代理人提供健康告知填写的专业指导，帮助代理人引导客户如实、正确地完成健康告知问卷，解析常见疑难问题的处理方式，避免因告知不当导致日后的理赔纠纷。
## 触发条件
当代理人或客户需要：
- 了解如何填写健康告知问卷
- 判断某种既往病史是否需要告知
- 处理健康告知中的疑难问题
- 理解如实告知的法律含义和重要性
## 前置条件insurance-agent-objection-handler
When your agent uses it
为保险代理人提供客户异议处理话术与策略的专业技能，帮助代理人有效应对常见拒绝理由，化解疑虑并推进成交。当代理人遇到客户提出保费太贵、不需要、再考虑考虑、没时间等异议，或在方案呈现后遭遇沉默、推拖时触发。适用于需求分析阶段的初期抵触、方案呈现后的价格异议、促成阶段的最后疑虑处理等场景。 触发词：太贵了、再考虑、不需要。
What it loads into context
Frontmatter
166 tokensAlways available to the agent
SKILL.md
3,189 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 输出格式
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset116 bytesSKILL.mdcore11,473 bytesconfig.jsonasset397 bytesrequirements.txtdoc137 bytesscripts/validate_input.pyruns1,603 bytes
Read SKILL.md
# 客户异议处理
## 角色定义
扮演保险销售顾问专家。为代理人提供客户异议处理的标准话术与策略，帮助代理人有效应对价格、需求、信任、时机等各类常见拒绝理由，化解客户疑虑并推进成交。
## 触发条件
当代理人遇到：
- 客户提出明确的拒绝理由（价格/不需要/没时间等）
- 客户沉默、拖延、不做决定
- 方案呈现后的疑虑和追问
- 促成环节的最后障碍
## 前置条件insurance-agent-plan-generator
When your agent uses it
保险计划书生成器。根据客户画像、保障缺口分析和产品组合推荐，自动生成专业Word(.docx)格式保险计划书。支持从客户画像技能和产品推荐技能导入数据，也支持独立手动录入。自动填充客户信息、产品方案、利益演示，生成合规精美的个性化计划书。当用户提到计划书、保险方案、方案书、规划书、生成计划书、制作方案、保障方案、保障规划时使用。 触发词：计划书、出方案、生成方案。
What it loads into context
Frontmatter
166 tokensAlways available to the agent
SKILL.md
3,687 tokensWhen the skill is activated
Reference documents
2,317 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 输出格式
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
lib/benefit_calculator.pylib/generate_plan.pyscripts/validate_input.py
Files
.env.exampleasset116 bytesSKILL.mdcore13,378 bytesTEST.mddoc2,186 bytesassets/data-input-form.htmlasset21,316 bytesassets/plan-config-form.htmlasset6,973 bytesconfig.jsonasset562 bytesdemo-data/demo_plan_input.jsonasset8,712 bytesexamples.mddoc4,371 byteslib/benefit_calculator.pyruns20,282 byteslib/generate_plan.pyruns33,064 bytesrequirements.txtdoc164 bytesscripts/validate_input.pyruns1,767 bytestemplates/benefit_models.jsonasset1,282 bytestemplates/compliance_text.jsonasset2,306 bytestemplates/plan_sections.mddoc1,644 bytes
Read SKILL.md
# 保险计划书生成器
## 角色定义
扮演保险计划书生成专家。根据客户画像、保障缺口分析和产品组合推荐，一键生成专业排版的Word格式保险计划书，支持技能联动、手动录入、文件导入三种数据输入模式。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 计划书
- 保险方案
- 方案书
- 保障规划
insurance-agent-policy-explanation
When your agent uses it
为保险代理人提供保单条款解读与产品讲解的专业指导。当需要向客户解释保险条款、解答保障范围疑问、处理理赔相关咨询时使用。帮助代理人用通俗语言解释复杂条款，建立客户信任。 触发词：条款解读、等待期、免责条款。
What it loads into context
Frontmatter
120 tokensAlways available to the agent
SKILL.md
2,617 tokensWhen the skill is activated
Reference documents
591 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 输出格式
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset116 bytesSKILL.mdcore9,200 bytesTEST.mddoc2,127 bytesconfig.jsonasset592 bytesrequirements.txtdoc139 bytesscripts/validate_input.pyruns1,599 bytes
Read SKILL.md
# Agent保单条款解读与产品讲解
## 角色定义
扮演保险条款解读与产品讲解专家。用通俗语言解释复杂保险条款，帮助客户理解保障范围、责任免除和理赔条件，建立专业信任。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 条款解读
- 保单讲解
- 保障范围
- 产品讲解
insurance-agent-product-wiki
When your agent uses it
保险代理人产品知识库与销售助手。上传产品文档自动入库，以LLM Wiki理念为每个产品建立独立知识页。集成学习问答、模拟演练、话术推荐三大模式。当用户提到保险产品、产品入库、话术推荐、模拟演练、学习产品、保险问答时使用此技能。 触发词：产品保障什么、话术演练、重疾和医疗区别。
What it loads into context
Frontmatter
143 tokensAlways available to the agent
SKILL.md
3,317 tokensWhen the skill is activated
Reference documents
41,817 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 输出格式
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
install.shscripts/extract_doc.pyscripts/test_skill_v2.pyscripts/validate_input.pyuninstall.sh
Files
.env.exampleasset116 bytesSKILL.mdcore11,950 bytesconfig.jsonasset471 bytesinstall.shruns5,602 bytesproduct-index.mddoc2,168 bytesproducts/guo-shou-fu-sheng-dian/raw/source-text.mddoc1,055 bytesproducts/guo-shou-fu-sheng-dian/scripts.mddoc2,918 bytesproducts/guo-shou-fu-sheng-dian/wiki.mddoc2,820 bytesproducts/hao-yi-bao-qi-xian-2025/raw/source-text.mddoc1,606 bytesproducts/hao-yi-bao-qi-xian-2025/scripts.mddoc3,406 bytesproducts/hao-yi-bao-qi-xian-2025/wiki.mddoc3,402 bytesproducts/jia-cheng-yi-wai-xian/raw/source-text.mddoc1,174 bytesproducts/jia-cheng-yi-wai-xian/scripts.mddoc2,841 bytesproducts/jia-cheng-yi-wai-xian/wiki.mddoc2,929 bytesproducts/ping-an-che-xian/raw/source-text.mddoc1,522 bytesproducts/ping-an-che-xian/scripts.mddoc2,663 bytesproducts/ping-an-che-xian/wiki.mddoc3,246 bytesproducts/ping-an-e-sheng-bao-yue-xiang/raw/source-text.mddoc1,496 bytesproducts/ping-an-e-sheng-bao-yue-xiang/scripts.mddoc2,827 bytesproducts/ping-an-e-sheng-bao-yue-xiang/wiki.mddoc3,720 bytesproducts/ping-an-e-sheng-fu-2024/raw/source-text.mddoc963 bytesproducts/ping-an-e-sheng-fu-2024/scripts.mddoc3,114 bytesproducts/ping-an-e-sheng-fu-2024/wiki.mddoc2,613 bytesproducts/ren-bao-che-xian/raw/source-text.mddoc927 bytesproducts/ren-bao-che-xian/scripts.mddoc2,501 bytesproducts/ren-bao-che-xian/wiki.mddoc2,852 bytesproducts/tai-kang-quan-neng-bao-2025/raw/source-text.mddoc1,471 bytesproducts/tai-kang-quan-neng-bao-2025/scripts.mddoc2,795 bytesproducts/tai-kang-quan-neng-bao-2025/wiki.mddoc3,433 bytesproducts/tai-ping-hong-fen-xian/raw/source-text.mddoc1,037 bytesproducts/tai-ping-hong-fen-xian/scripts.mddoc2,692 bytesproducts/tai-ping-hong-fen-xian/wiki.mddoc2,789 bytesproducts/zun-xiang-e-sheng-2025/raw/source-text.mddoc1,944 bytesproducts/zun-xiang-e-sheng-2025/scripts.mddoc2,839 bytesproducts/zun-xiang-e-sheng-2025/wiki.mddoc3,799 bytesrequirements.txtdoc132 bytesschema.mddoc3,718 bytesscripts/extract_doc.pyruns5,570 bytesscripts/test_skill_v2.pyruns11,530 bytesscripts/validate_input.pyruns1,661 bytestemplates/product_wiki.mddoc2,010 bytestemplates/script_template.mddoc3,698 bytestest-report-v2.jsonasset7,019 bytestest-report.mddoc44,282 bytesuninstall.shruns1,259 bytes
Read SKILL.md
# 保险产品Wiki与销售助手
## 角色定义
扮演保险产品专家与销售教练。为保险代理人构建产品知识库，提供产品入库、学习问答、模拟演练、话术推荐四大核心能力，帮助代理人把产品条款转化为客户听得懂的销售语言。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 产品知识库
- 产品入库
- 话术推荐
- 模拟演练
insurance-agent-renewal-assist
When your agent uses it
为保险代理人提供保单续期提醒、续期维护话术和客户防流失策略的专业技能，帮助代理人提高客户续期率，减少因欠费导致的保单失效。当客户保单即将到期需要续期、客户迟迟未缴纳续期保费、需要做续期前保单健康检查时触发。适用于续期前30天提醒、欠费催缴、考虑退保时的挽留，以及续期时借机加保等场景。 触发词：保单到期、续期话术、退保挽留。
What it loads into context
Frontmatter
182 tokensAlways available to the agent
SKILL.md
2,949 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 输出格式
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset116 bytesSKILL.mdcore10,446 bytesconfig.jsonasset758 bytesrequirements.txtdoc134 bytesscripts/validate_input.pyruns1,938 bytes
Read SKILL.md
# 续期维护与防流失
## 角色定义
扮演保险续期维护与防流失专家。
## 触发条件
当代理人需要：
- 提醒客户即将到来的续期保费
- 跟进已欠费的保单避免失效
- 挽留考虑退保的客户
- 借助续期时机推动加保
## 前置条件insurance-agent-social-media
When your agent uses it
扮演保险社交媒体营销专家。为保险代理人自动获取热点新闻，生成高质量内容，一键确认后通过Playwright自动发布，帮助代理人打造专业且有温度的个人IP。 触发词：保险热点、营销文案、小红书笔记。
What it loads into context
Frontmatter
106 tokensAlways available to the agent
SKILL.md
4,266 tokensWhen the skill is activated
Reference documents
2,936 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 输出格式
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.pyscripts/xiaohongshu_publish.py
Files
.env.exampleasset116 bytesSKILL.mdcore15,294 bytesSKILL.mdcore15,294 bytesconfig.jsonasset686 bytescontent-templates.mddoc6,650 bytesimage-templates.mddoc3,753 bytesrequirements.txtdoc146 bytesscripts/validate_input.pyruns1,946 bytesscripts/xiaohongshu_publish.pyruns12,209 bytes
Read SKILL.md
# 保险代理人社交媒体智能营销助手
## 角色定义
扮演保险社交媒体营销专家。为保险代理人自动获取热点新闻，生成高质量内容，一键确认后通过Playwright自动发布，帮助代理人打造专业且有温度的个人IP。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 小红书
- 保险营销
- 保险推广
- 营销笔记
- 线上获客a-market-hotspot-discovery
When your agent uses it
面向A股市场热点发现的综合能力。适用于用户提出“今日热点是什么”“最近哪些板块最热”“当前热股有哪些”等泛化问题时，返回结构化的Markdown热点报告。覆盖热点资讯、热门板块、热门股票等维度。触发核心条件：用户关注市场层面的热门方向或事件热度，而非单一标的的深度诊断或指标计算。数据查询请使用 gildata-aidata 服务。
What it loads into context
Frontmatter
111 tokensAlways available to the agent
SKILL.md
1,406 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 核心触发条件
- 数据工具集 (gildata-aidata)
- 工作流程
- 报告输出模板
- 使用示例
- 注意事项
Files
SKILL.mdcore4,875 bytes
Read SKILL.md
# A股市场热点发现技能
## 概述
本技能专注于**A股市场盘面热点与资金情绪监测**。当用户询问市场整体热门方向、资金聚集地、突发题材或个股异动时，通过整合板块涨幅、资金流向、龙虎榜数据及全网舆情，输出结构化的《A股市场热点监测报告》。
## 核心触发条件
- 用户询问“今日市场热点”、“今天什么板块涨得好”、“最近炒什么概念”。
- 用户关注“资金在买什么”、“龙虎榜有哪些大佬进场”、“今日热门股”。
- 用户需要了解“盘面情绪”、“题材轮动”、“政策/事件驱动方向”。
## 数据工具集 (gildata-aidata)
| 工具名称 | 核心作用 | 关键参数 |
|----------|----------|----------|comparable-company-analysis
When your agent uses it
金融领域的可比公司分析工具，通过行业分析法对目标公司与可比公司整合经营指标、估值倍数和统计基准比较。支持A股、港股、美股市场的上市公司分析。输入单一公司问句后，调用可比公司接口并生成主题化Excel报告。适用于快速对比目标公司与同业在经营、财务、估值维度的差异。数据查询优先使用 gildata-aidata 服务。
What it loads into context
Frontmatter
112 tokensAlways available to the agent
SKILL.md
1,259 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 核心触发条件
- 数据工具集 (gildata-aidata)
- 工作流程
- 报告输出结构 (Excel)
- 使用示例
- 注意事项
Files
SKILL.mdcore4,589 bytes
Read SKILL.md
# 可比公司分析技能 (Comparable Company Analysis)
## 概述
本技能专注于**上市公司同业对比与估值定位**。当用户提供单一目标公司时，自动通过行业分析法筛选可比公司，整合经营指标、财务比率与估值倍数，生成标准化的《可比公司分析 Excel 报告》，辅助快速判断目标公司在同业中的相对地位与估值高低。
## 核心触发条件
- 用户询问“帮我做个XX公司的可比公司分析”、“XX公司和同行比怎么样”、“XX公司估值贵不贵”。
- 用户需要快速对比目标公司与同业在经营、财务、估值维度的差异。
- 涉及跨市场（A股、港股、美股）上市公司的横向对比需求。
## 数据工具集 (gildata-aidata)
| 工具名称 | 核心作用 | 关键参数 |
|----------|----------|----------|fund-diagnosis
When your agent uses it
公募基金单基金综合诊断能力。适用于用户提出"这只基金怎么样""适不适合继续持有""风险和收益特征如何""帮我看看这只基金""XX基金好不好""XX基金值得买吗""XX基金最近表现如何""帮我分析一下XX基金"等泛化诊断问题时，返回结构化的Markdown诊断报告。 当用户询问某只具体基金的整体评价、持有建议、风险收益特征、基金经理能力、持仓结构、业绩表现等综合性问题时必须使用此技能。 即使用户没有明确说"诊断""分析"，只要涉及单基金的综合性评价就应触发。 覆盖场景：基金评价、持有建议、风险评估、业绩解读、经理分析、持仓解读、基金体检等。 每次仅分析一只基金，不处理多基金对比与量化建模。 不适用于：多基金对比、基金筛选、具体指标计算、回测建模等场景（这些场景请使用其他对应技能）。
What it loads into context
Frontmatter
254 tokensAlways available to the agent
SKILL.md
2,957 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能定位
- 触发场景
- 诊断流程
- 报告结构
- 数据解读指南
- 注意事项
- 工具调用示例
Files
SKILL.mdcore10,153 bytesSKILL.mdcore10,153 bytes
Read SKILL.md
# 公募基金单基金综合诊断
## 技能定位
本技能专注于**单只公募基金**的全面诊断分析，通过多维度数据整合，为用户提供结构化的基金评估报告。
**核心原则：每次仅分析一只基金，不处理多基金对比。**
## 触发场景
当用户提出以下类型的问题时触发本技能：
- "这只基金怎么样？"
- "XX基金适不适合继续持有？"fund-multi-factor-filter
When your agent uses it
通过自然语言查询进行公募基金筛选技能。根据基金类型、业绩、基金经理、风险、持仓、资产配置等多维度组合筛选公募基金。返回符合条件的相关基金数据。当用户询问基金筛选问题时，必须使用此技能。
What it loads into context
Frontmatter
73 tokensAlways available to the agent
SKILL.md
1,303 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 数据工具集 (gildata-aidata)
- 核心筛选维度
- 工作流程
- 报告输出模板
- 使用示例
Files
SKILL.mdcore4,436 bytes
Read SKILL.md
# 公募基金智能选基技能
## 概述
本技能专注于**多维度公募基金筛选**。通过整合业绩表现、风险控制、基金经理能力、资产配置、行业偏好等数据，帮助用户快速从海量公募基金中精准定位符合特定投资逻辑的标的。
## 数据工具集 (gildata-aidata)
| 工具名称 | 核心作用 | 关键参数 |
|----------|----------|----------|
| `FundMultipleFactorFilter` | **智能选基** | 接收自然语言查询，返回符合多条件筛选的基金列表 |
## 核心筛选维度
### 1. 业绩与收益指标stock-fund-analysis
When your agent uses it
股票资金流向分析技能。专业诊断主力资金流向、北向/机构资金、量能换手率、资金趋势与筹码集中度四大核心维度，客观判断资金认可度与进出意图。当用户询问股票资金流向、主力资金、北向资金、机构持仓、筹码分布、量价关系、资金面分析、个股资金诊断时使用。即使用户没有明确说"资金分析"，只要涉及股票资金面研判就应触发。
What it loads into context
Frontmatter
126 tokensAlways available to the agent
SKILL.md
2,988 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 四大核心分析维度
- 分析流程
- 注意事项
- 数据获取优先级
Files
SKILL.mdcore10,248 bytes
Read SKILL.md
# 股票资金分析技能
## 概述
本技能专注于**股票资金面分析**，通过四大核心维度对个股或板块的资金动向进行系统性诊断，帮助判断主力资金意图与市场认可度。
**重要原则：本技能仅提供资金面客观数据分析，不构成任何投资建议。所有分析结论均需结合基本面、技术面、消息面综合判断。**
## 四大核心分析维度
### 维度一：主力资金流向诊断
追踪大资金进出轨迹，识别主力控盘程度与操作意图。
stock-multi-factor-filter
When your agent uses it
通过自然语言查询进行 A 股股票筛选技能，支持行情指标、技术形态、财务指标、行业概念等多条件组合筛选。返回符合条件的相关股票数据。当用户询问针对行情、财务数据、技术指标、行业概念等 A 股股票筛选相关问题时，必须使用此技能。
What it loads into context
Frontmatter
84 tokensAlways available to the agent
SKILL.md
1,208 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 数据工具集 (gildata-aidata)
- 核心筛选维度
- 工作流程
- 报告输出模板
- 使用示例
Files
SKILL.mdcore3,923 bytes
Read SKILL.md
# A 股智能选股技能
## 概述
本技能专注于**多维度 A 股股票筛选**。通过整合市场行情、技术形态、财务基本面、行业概念及资金流向等数据，帮助用户快速从数千只股票中精准定位符合特定投资逻辑的标的。
## 数据工具集 (gildata-aidata)
| 工具名称 | 核心作用 | 关键参数 |
|----------|----------|----------|
| `StockMultipleFactorFilter` | **智能选股** | 接收自然语言查询，返回符合多条件筛选的股票列表 |
## 核心筛选维度
### 1. 行情与量价指标stock-quote-analysis
When your agent uses it
股票行情与技术分析技能。基于 gildata-aidata 服务查询个股实时行情、分时走势、量价表现与技术指标。当用户询问“今天表现如何”、“近期走势如何”、“是否放量”、“技术面怎么看”、“当前股价多少”或查询 K 线形态、均线、MACD 等技术指标时使用。即使用户未明确说“行情分析”，只要涉及股价走势研判就应触发。
What it loads into context
Frontmatter
107 tokensAlways available to the agent
SKILL.md
1,799 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 核心工具集 (优先使用 gildata-aidata)
- 分析维度与逻辑
- 工作流程
- 报告输出模板
- 注意事项
Files
SKILL.mdcore5,877 bytes
Read SKILL.md
# 股票行情与技术分析技能
## 概述
本技能专注于**股票行情与技术面分析**，利用权威数据源解读个股的量价表现、趋势特征及技术信号。适用于回答用户关于“今天表现如何”、“近期走势”、“量价关系”、“技术指标状态”等问题。
**重要原则：本技能仅提供基于数据的客观行情解读，不构成买卖建议。技术分析具有滞后性和概率性，需结合市场环境综合判断。**
## 核心工具集 (优先使用 gildata-aidata)
| 工具名称 | 功能描述 | 适用场景 |
|----------|----------|----------|
| `AShareLiveQuote` | A股实时行情 | 查询当前股价、涨跌幅、成交量、换手率、市值等 |
| `AShareTickQuote` | A股分时行情 | 查询指定日期的分时走势（价量变化），判断盘中资金博弈 |
| `StockDailyQuote` | 股票日行情 | 查询指定日期的日 K 线数据 |stock-shareholder-analysis
When your agent uses it
公司股东股本查询分析技能。查询股本结构、股权结构、股东户数、前十大股东/流通股东、主要持有人、实控人等股权信息，支持自然语言问句输入，返回相关股东股本数据结果。当用户询问股本结构、股东户数、前十大股东、股权质押、实控人、主要持有人等股东股本数据查询问题时，必须使用此技能。
What it loads into context
Frontmatter
114 tokensAlways available to the agent
SKILL.md
1,634 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 数据工具集 (gildata-aidata)
- 核心分析维度
- 工作流程
- 报告输出模板
Files
SKILL.mdcore5,460 bytes
Read SKILL.md
# 公司股东股本查询分析技能
## 概述
本技能专注于**上市公司股东与股本结构**的深度查询与分析。通过整合股本数据、股东名单、持股变动及质押风险等信息，帮助用户快速透视一家公司的**股权架构**与**筹码分布**特征，辅助判断公司治理质量及潜在的投资风险/机会。
## 数据工具集 (gildata-aidata)
| 工具名称 | 核心作用 | 关键参数 |
|----------|----------|----------|
| `StockShareStructure` | **股本结构** | 获取总股本、流通股本、限售股及变动原因 |
| `Top10ShareHolders` | **十大股东** | 获取前十大股东名称、持股数量及比例 |
| `Top10FloatShareHolders` | **十大流通股东** | 获取前十大流通股东名称、持股数量及比例 |
| `ShareholderNum` | **股东户数** | 获取股东户数变化、户均持股，分析筹码集中度 |
| `StockPledge` | **股权质押** | 获取股东质押明细，评估股权爆仓风险 |stock-tech-analysis
When your agent uses it
股票深度技术分析技能。专业诊断趋势、量价、支撑压力、技术指标四大核心维度，结合个股市值规模（盘子大小）与行业属性，客观判断股价走势强弱与关键交易区间。当用户需要深度技术面诊断、寻找买卖点参考、分析趋势强度时使用。
What it loads into context
Frontmatter
86 tokensAlways available to the agent
SKILL.md
1,935 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 数据工具集 (gildata-aidata)
- 核心分析维度
- 结合个股特征的修正逻辑
- 工作流程
- 报告输出模板
Files
SKILL.mdcore6,206 bytes
Read SKILL.md
# 股票深度技术分析技能
## 概述
本技能专注于**股票技术面的深度诊断**，超越基础的行情播报，提供基于多维度的逻辑研判。通过综合分析趋势结构、量价配合、关键位及技术指标，并结合**流通盘大小**和**行业特征**进行修正，为用户提供客观的走势强弱评估和关键交易区间参考。
## 数据工具集 (gildata-aidata)
| 工具名称 | 核心作用 | 关键参数 |
|----------|----------|----------|
| `AShareLiveQuote` | **实时行情与市值** | 获取现价、换手率、量比、总市值、流通市值（判断盘子大小） |
| `StockQuoteTechIndex` | **技术指标全览** | 获取 MA、MACD、KDJ、RSI、BOLL 指标及支撑/压力位数值 |
| `StockRangeQuotation` | **区间统计** | 计算近期高低点、振幅、成交额，辅助判断支撑压力 |
| `StockMultiPeriodQuote` | **多周期共振** | 日/周/月线趋势对比，判断大级别方向 |
When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
2,737 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 核心能力
- 工作流程
- 报告结构
- 工具调用规范
- 注意事项
Files
SKILL.mdcore9,191 bytesSKILL.mdcore9,191 bytesSKILL.mdcore9,191 bytesSKILL.mdcore9,191 bytes
Read SKILL.md
# 另类因子挖掘技能
你是证券金融工程研究员，专注于另类数据（Alternative Data）的因子挖掘与Alpha来源探索。
## 核心能力
本技能围绕**另类数据→因子构建→有效性检验→超额归因**的完整链路，系统化挖掘传统量价、财务因子之外的超额收益来源。
## 工作流程
### 第一阶段：另类数据获取与萃取
根据用户需求，确定另类数据类型并通过 `gildata-aidata` 服务获取数据：
1. **舆情数据**：使用 `finx gildata-aidata StockNewslist` 获取目标股票相关舆情，使用 `finx gildata-aidata NewsInfoList` 获取全网舆情，使用 `finx gildata-aidata IndustryNewslist` 获取行业舆情announcement-analysis
When your agent uses it
解读上市公司重大公告（年报、季报、业绩预告、并购重组、定增、增减持、股权激励、重大合同等），生成结构化的解读报告。适用于基金研究员、投资经理等需要快速理解公告核心信息、分析预期差、提炼投资要点的场景。当用户提到公告解读、公告分析、公告点评、研报点评、业绩解读、财报分析、定增分析、减持分析、股权激励分析、重大合同分析、并购重组分析、回购分析、股权质押分析、诉讼处罚分析时使用此 skill。默认输出结构化 Markdown 格式报告。
What it loads into context
Frontmatter
163 tokensAlways available to the agent
SKILL.md
8,039 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能定位
- 触发场景
- 工作流程
- 报告结构（Markdown 格式）
- 格式规范
- 专项分析模板
- 分析质量要求
- 工具使用指引
- 注意事项
- gildata-aidata 工具调用指南
- 数据来源优先级
- 技能更新日志
Files
SKILL.mdcore27,782 bytesSKILL.mdcore27,782 bytesSKILL.mdcore27,782 bytesSKILL.mdcore27,782 bytesSKILL.mdcore27,782 bytes
Read SKILL.md
# 重大公告解读技能
## 技能定位
本技能帮助基金行业研究员、投资经理等专业人士快速、专业地解读上市公司重大公告，**以 gildata-aidata 金融数据服务为核心数据源**，生成结构化的解读报告。核心方法论采用券商研报的标准点评框架，注重边际变化提取和预期差分析。
**核心优势**：
- ✅ **gildata-aidata 专业数据源**：所有财务数据、估值数据、一致预期优先通过 gildata-aidata 服务获取
- ✅ **分析框架专业**：采用券商研报标准点评框架，注重边际变化和预期差分析
- ✅ **覆盖场景全面**：支持 13 种公告类型的专项分析模板
- ✅ **输出结构化**：标准化 Markdown 格式报告，适合投资决策参考
**输出格式**：
- 默认生成结构化 Markdown 格式报告company-deep-analysis
When your agent uses it
上市公司深度分析技能。当用户要求对公司进行深度分析、全面研究、投资价值分析时使用，包括"分析 XX 公司"、"对 XX 公司做深度研究"、"研究 XX 公司的投资价值"、"XX 公司深度分析"、"XX 公司全面分析"、"帮我深度研究一下 XX 公司"等表述。本技能从财务分析、业务分析、行业分析、估值分析、风险分析、管理层分析、政策风险七大维度，利用东方财富妙想、企业工商查询等工具获取权威数据，生成面向专业研究员的结构化 Markdown 深度研究报告。**报告生成后直接在对话中显示完整内容，无需发送文件**。
What it loads into context
Frontmatter
168 tokensAlways available to the agent
SKILL.md
5,699 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能定位
- 触发场景
- 执行流程
- 报告结构模板
- 特殊场景处理
- 工具使用说明
- gildata-aidata 工具调用指南
- 数据来源优先级
- 质量检查清单
- 注意事项
- 技能更新日志
Scripts executed separately
output/五粮液深度分析报告.mdoutput/贵州茅台深度分析报告.md
Files
SKILL.mdcore19,239 bytesSKILL.mdcore19,239 bytesSKILL.mdcore19,239 bytesSKILL.mdcore19,239 bytesSKILL.mdcore19,239 bytesoutput/五粮液深度分析报告.mdruns21,981 bytesoutput/贵州茅台深度分析报告.mdruns22,054 bytes
Read SKILL.md
# 上市公司深度分析技能
## 技能定位
本技能面向**专业研究员**，提供上市公司全方位深度分析。通过调用 `gildata-aidata` 专业金融数据服务与 `asset-service` 企业工商查询工具，从七大维度系统性地分析公司投资价值，生成机构级深度研究报告。
**核心优势**：
- ✅ **gildata-aidata 专业数据源**：所有财务数据、估值数据、一致预期优先通过 gildata-aidata 服务获取
- ✅ **分析框架系统**：覆盖财务、业务、行业、估值、风险、管理层、政策七大维度
- ✅ **深度逻辑推演**：不仅罗列数据，更注重数据背后的商业逻辑与投资价值
- ✅ **输出结构化**：标准化 Markdown 格式报告，直接在对话中展示完整内容
**输出格式**：
- 默认生成结构化 Markdown 格式报告company-one-page-analysis
When your agent uses it
上市公司一页纸分析技能。当用户要求对公司进行简洁分析、快速点评、一页纸分析、速评时使用，包括"XX 公司一页纸"、"快速分析 XX 公司"、"XX 公司速评"、"XX 公司简单分析"、"XX 公司核心要点"等表述。本技能从核心财务、业务亮点、估值水平、投资逻辑、主要风险五大维度，利用 gildata-aidata 专业金融数据服务获取核心财务与估值数据，生成简洁的一页纸 Markdown 分析报告，聚焦核心信息，适合快速决策参考。**报告生成后直接在对话中显示完整内容，无需发送文件**。
What it loads into context
Frontmatter
164 tokensAlways available to the agent
SKILL.md
4,052 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能定位
- 触发场景
- 数据获取策略
- 分析框架（一页纸五大维度）
- 报告输出格式
- 执行流程
- gildata-aidata 工具调用指南
- 数据来源优先级
- 注意事项
- 特殊情况处理
- 示例调用
- 与深度分析技能对比
- 技能更新日志
Files
SKILL.mdcore13,784 bytesSKILL.mdcore13,784 bytesSKILL.mdcore13,784 bytesSKILL.mdcore13,784 bytes
Read SKILL.md
# 上市公司一页纸分析技能
## 技能定位
本技能面向**投资经理、研究员、基金经理**，提供上市公司快速分析工具。在一页纸篇幅内，聚焦公司最核心的投资信息，帮助快速判断公司投资价值和风险，适合晨会汇报、快速决策、标的初筛等场景。
**核心优势**：
- ✅ **gildata-aidata 专业数据源**：所有财务数据、估值数据优先通过 gildata-aidata 服务获取
- ✅ **精炼高效**：五大核心维度，一页纸篇幅，直击投资要点
- ✅ **快速决策**：适合晨会汇报、标的跟踪、快速初筛
与 `company-deep-analysis` 的区别：
- **深度分析**：七大维度，全面深入，适合深度研究和首次覆盖
- **一页纸分析**：五大核心维度，精炼简洁，适合快速决策和标的跟踪convertible-bond-valuation
When your agent uses it
证券固定收益研究员转债估值与博弈速评技能。当用户要求分析可转债、转债估值、转股溢价率、强赎风险、下修风险、转债套利机会时使用。包括"分析XX转债"、"转债估值速评"、"测算转股溢价率"、"强赎风险分析"、"下修博弈分析"、"转债套利机会"、"转债投资价值分析"、"可转债点评"、"转债排雷"等表述。即使用户只说"看看这个转债"、"转债怎么样"、"分析下XX转债"等简短表述也要触发。本技能自动获取转债行情数据、正股信息、溢价率指标，测算强赎/下修条件，挖掘套利机会，生成标准化Markdown格式转债估值与博弈速评报告。
What it loads into context
Frontmatter
227 tokensAlways available to the agent
SKILL.md
3,393 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能概述
- 数据获取策略
- 工作流程
- 工具使用规范
- 注意事项
- 示例调用
- 技能更新日志
Files
SKILL.mdcore10,834 bytes
Read SKILL.md
# 转债估值与博弈速评
## 技能概述
本技能面向证券固定收益研究员，提供可转债的快速估值分析与博弈机会挖掘。核心功能包括转股溢价率测算、强赎风险评估、下修博弈分析、套利机会识别，输出标准化Markdown格式速评报告。
## 数据获取策略
### 核心数据源：gildata-aidata
所有转债行情、正股数据、公告舆情均优先通过 `gildata-aidata` 服务获取。
| 数据需求 | 推荐工具 | 说明 |
|---------|---------|------|When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
6,624 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能概述
- 触发场景
- 工作流程
- 输出模板
- 特殊场景处理
- 输出后处理
- 使用示例
- 注意事项
- 数据来源优先级
- 工具使用说明
- gildata-aidata 工具调用指南
Files
SKILL.mdcore22,200 bytesSKILL.mdcore22,200 bytesSKILL.mdcore22,200 bytesSKILL.mdcore22,200 bytesSKILL.mdcore22,200 bytes
Read SKILL.md
# 上市公司财报业绩点评报告生成技能
## 技能概述
本技能专为上市公司财报（年报/半年报/季报）分析设计，**以 gildata-aidata 金融数据服务为核心数据源**，生成结构化业绩点评报告。报告包含业绩概览、核心财务指标分析、业务板块拆解、盈利质量评估、风险提示及投资观点参考，帮助投资者快速把握公司经营状况。
**核心优势**：
- ✅ **gildata-aidata 专业数据源**：所有财务数据优先通过 gildata-aidata 服务获取，确保数据权威准确
- ✅ **数据全面**：覆盖营收、利润、毛利率、现金流、ROE 等核心指标
- ✅ **分析专业**：采用买方研究员分析框架，识别经营亮点与风险点
- ✅ **报告结构化**：8 大模块标准化输出，适合投资决策参考
---
## 触发场景fixed-income-daily-brief
When your agent uses it
固定收益研究员日报生成技能。当用户要求生成固收日报、债券市场日报、债市复盘、利率日报、资金面日报时使用，包括"固收日报"、"债市日报"、"债券日报"、"利率市场日报"、"资金面日报"、"生成固收日报"、"今日债市复盘"、"固收日报速评"、"债市每日简报"等表述。本技能自动获取资金面、债券市场、央行操作、信用债、同业存单、地方债、二级资本债、永续债、ABS等核心数据，结合市场情绪与政策动态，生成标准化Markdown格式固定收益日报。即使用户只说"今天债市怎么样"、"生成债券日报"、"固收日报"等简短表述也要触发。
What it loads into context
Frontmatter
202 tokensAlways available to the agent
SKILL.md
3,965 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能概述
- 触发场景
- 数据获取流程
- 报告结构
- 数据获取注意事项
- 数据获取失败处理
- 输出要求
- 高级功能
- 技能调用示例
Files
SKILL.mdcore12,865 bytesSKILL.mdcore12,865 bytesSKILL.mdcore12,865 bytesSKILL.mdcore12,865 bytesSKILL.mdcore12,865 bytesSKILL.mdcore12,865 bytesSKILL.mdcore12,865 bytesSKILL.mdcore12,865 bytesSKILL.mdcore12,865 bytesSKILL.mdcore12,865 bytesSKILL.mdcore12,865 bytesSKILL.mdcore12,865 bytesSKILL.mdcore12,865 bytes
Read SKILL.md
# 固定收益日报生成技能
## 技能概述
本技能面向证券固定收益研究员，自动生成标准化的固定收益市场日报。通过 `gildata-aidata` 专业金融数据服务获取权威数据，覆盖资金面、利率债、信用债、同业存单、地方债、二级资本债、永续债、ABS、央行操作及重要政策舆情，输出专业、结构化的Markdown格式日报。
## 触发场景
- 用户要求生成固收日报、债市日报、债券日报、利率日报、资金面日报
- 用户询问"今天债市情况"、"资金面如何"、"债券市场复盘"
- 用户需要每日固定收益市场回顾与展望
- 用户要求生成固收晨报、债市早报、固收晚报
- 即使用户表述简短如"固收日报"、"债市日报"、"今天债券市场"也要触发
When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
3,332 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能定位
- 触发场景
- 执行流程
- 数据获取指引
- 注意事项
- 输出要求
Files
SKILL.mdcore11,190 bytes
Read SKILL.md
# 固定收益隔夜舆情速览
## 技能定位
本技能专为证券固定收益研究员设计，用于快速聚合隔夜及最新债券市场相关舆情，生成每日雷点清单和速览报告，帮助研究员在开盘前掌握市场动态、识别信用风险、跟踪政策变化。
## 触发场景
- 用户要求生成隔夜固收舆情速览、固收晨报、债市日报
- 用户询问今日债市有什么风险点、雷点、需要关注的事件
- 用户要求跟踪债券违约、评级下调、评级展望负面等信用事件
- 用户要求汇总央行公开市场操作、MLF、LPR、降准降息等货币政策动态
- 用户要求梳理最新宏观政策、财政政策、监管政策对债市的影响
- 用户提及"固收舆情"、"债券市场要闻"、"信用债跟踪"、"利率债动态"等关键词
When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
2,729 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能概述
- 数据获取流程
- 数据处理与分析
- 报告生成
- 数据获取指引
- 执行指引
- 注意事项
Files
SKILL.mdcore8,330 bytes
Read SKILL.md
# 固定收益研究员资金面&利率复盘简报
## 技能概述
本技能面向证券固定收益研究员，自动化生成资金面与利率复盘简报。通过获取银行间市场核心利率数据、分析利率变动逻辑、结合宏观与政策因素，输出结构化的复盘报告。
## 数据获取流程
### 第一步：获取当前日期
使用 `finx time` 获取当前日期，确定复盘的交易日。
### 第二步：获取资金面核心数据
使用 `finx gildata-aidata` 查询以下数据：fixed-income-weekly-brief
When your agent uses it
固定收益研究员周报生成技能。当用户要求生成固收周报、债市周报、债券周报、利率周报、资金面周报时使用，包括"固收周报"、"债市周报"、"债券周报"、"利率周报"、"生成本周固收周报"、"本周债市复盘"、"固收周报速评"、"债市每周简报"、"本周债券市场回顾"等表述。本技能自动获取资金面、债券市场、央行操作、信用债、同业存单、地方债、二级资本债、永续债、ABS等核心周度数据，结合市场情绪与政策动态，生成标准化Markdown格式固定收益周报。即使用户只说"这周债市怎么样"、"生成债券周报"、"固收周报"等简短表述也要触发。
What it loads into context
Frontmatter
225 tokensAlways available to the agent
SKILL.md
6,109 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能概述
- 数据获取流程
- 数据处理与分析
- 报告结构
- 数据获取注意事项
- 数据获取失败处理
- 输出要求
- 高级功能
- 技能调用示例
Files
SKILL.mdcore19,467 bytesSKILL.mdcore19,467 bytesSKILL.mdcore19,467 bytesSKILL.mdcore19,467 bytesSKILL.mdcore19,467 bytesSKILL.mdcore19,467 bytesSKILL.mdcore19,467 bytesSKILL.mdcore19,467 bytesSKILL.mdcore19,467 bytesSKILL.mdcore19,467 bytesSKILL.mdcore19,467 bytesSKILL.mdcore19,467 bytesSKILL.mdcore19,467 bytes
Read SKILL.md
# 固定收益周报生成技能
## 技能概述
本技能面向证券固定收益研究员，自动生成标准化的固定收益市场周报。通过 `gildata-aidata` 专业金融数据服务获取权威数据，覆盖一周内的资金面、利率债、信用债、同业存单、地方债、二级资本债、永续债、ABS、央行操作及重要政策舆情，输出专业、结构化的Markdown格式周报。
与日报不同，周报更侧重于**周度趋势的确认**、**逻辑的连贯性**以及**下周的前瞻性**，强调将一周高频数据串联成债市叙事，并给出明确的配置建议。
## 数据获取流程
### 第一步：获取当前日期
使用 `time` 工具获取当前日期，确定报告覆盖的周期（通常为周一至周五或最近五个交易日）。
global-finance-brief
When your agent uses it
全球财经要闻自动抓取与简报生成技能。当用户询问财经热点、财经新闻、市场动态、投资热点、生成财经简报，或要求查看今天/最近金融市场有什么大事时触发。即使用户只说"财经热点"、"财经新闻汇总"等简短表述也要触发。本技能自动搜索最新财经资讯（24-72 小时），智能筛选高影响力事件，按宏观/市场/行业/公司分类汇总，生成结构化简要报告，包含核心头条、市场表现、行业动态、投资机会和风险提示。支持定时任务创建。
What it loads into context
Frontmatter
148 tokensAlways available to the agent
SKILL.md
3,686 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能定位
- 触发场景
- 执行流程
- 工具使用说明
- 质量检查清单
- 注意事项
Files
SKILL.mdcore12,581 bytesSKILL.mdcore12,581 bytesSKILL.mdcore12,581 bytes
Read SKILL.md
# 全球财经要闻简报技能
## 技能定位
本技能用于**自动抓取全球最新财经要闻**，通过 `gildata-aidata` 金融数据服务获取结构化数据与舆情，智能筛选高影响力事件，生成**结构化简要总结报告**。适用于用户快速获取每日财经热点、市场动态和投资机会参考。
**核心特点**：
- 自动捕获最新财经资讯（24-72 小时内）
- 智能筛选高影响力事件
- 结构化分类汇总
- 简洁易懂的简报格式
---
global-macro-linkage
When your agent uses it
海外宏观联动分析技能。当用户要求分析海外宏观数据、美联储政策、美国经济对国内影响、中美联动、外资流向、汇率波动对 A 股债市影响时使用。包括“分析美联储最新决议影响”、“美国经济衰退对中国影响”、“中美宏观联动分析”、“海外宏观周报”、“外资流向分析”、“汇率波动对资产影响”等表述。本技能自动跟踪海外核心宏观指标与政策动态，分析其通过汇率、贸易、情绪、通胀等渠道对国内经济和资本市场的传导影响，输出结构化 Markdown 联动分析简报。
What it loads into context
Frontmatter
160 tokensAlways available to the agent
SKILL.md
3,161 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 数据获取策略
- 执行流程
- 注意事项
- 示例调用
- 输出要求
- 技能更新日志
Files
SKILL.mdcore11,023 bytes
Read SKILL.md
# 海外宏观联动分析技能
## 概述
本技能模拟证券宏观研究员的海外研究视角，自动抓取全球核心宏观数据与政策动态，深入剖析海外经济对中国宏观经济、货币政策及资本市场的多维传导路径，输出专业的**海外 - 国内宏观联动分析简报**。
## 数据获取策略
### 核心数据源：gildata-aidata
所有宏观数据、政策、资金流向、行情及舆情数据均优先通过 `gildata-aidata` 服务获取。
| 数据需求 | 推荐工具 | 说明 |
|---------|---------|------|When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
5,320 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能概述
- 工作流程
- 风控报告模板
- 数据获取指引
- 风险信号判定规则
- 注意事项
- 输出要求
Files
SKILL.mdcore17,166 bytes
Read SKILL.md
# 产业债财报风控体检
## 技能概述
本技能面向证券固定收益研究员的产业债投资与风控场景，提供发债主体全方位财务风险筛查能力。通过自动获取企业财务数据、司法风险、评级变动、担保信息等，系统性识别现金流恶化、有息负债压力、担保链传染、诉讼纠纷等负面信号，输出结构化风控体检报告，辅助债券投资决策与持仓风险管理。
## 工作流程
### 第一步：确认分析主体
获取用户指定的企业名称或债券代码。若用户提供的是债券代码，先通过 `finx gildata-aidata CreditBondBaseInfo` 查询债券对应的发债主体名称。
若主体为非上市公司，需明确告知用户财务数据获取可能受限，将主要依赖工商信息、司法信息、评级信息等进行风险评估。
### 第二步：获取企业基础信息industry-deep-analysis
When your agent uses it
行业深度分析技能。当用户要求对某个行业进行深度分析、全面研究、行业投资价值分析时使用，包括"分析 XX 行业"、"对 XX 行业做深度研究"、"XX 行业投资价值分析"、"XX 行业深度分析"、"XX 行业全面分析"、"帮我深度研究一下 XX 行业"等表述。本技能从行业概况、市场规模、产业链分析、竞争格局、行业趋势、进入壁垒、风险因素、投资机会八大维度，利用 gildata-aidata 专业金融数据服务获取权威数据，生成面向专业研究员的结构化 Markdown 行业深度研究报告。**报告生成后直接在对话中显示完整内容，无需发送文件**。
What it loads into context
Frontmatter
181 tokensAlways available to the agent
SKILL.md
5,729 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能定位
- 触发场景
- 数据获取策略
- 分析框架（八大维度）
- 报告输出格式
- 执行流程
- gildata-aidata 工具调用指南
- 数据来源优先级
- 注意事项
- 特殊情况处理
- 示例调用
- 技能更新日志
Files
SKILL.mdcore19,865 bytesSKILL.mdcore19,865 bytes
Read SKILL.md
# 行业深度分析技能
## 技能定位
本技能面向**专业研究员、投资机构、产业分析师**，提供全方位行业深度分析。通过调用多个专业金融数据工具和行业数据库，从八大维度系统性地分析行业投资价值、竞争格局和发展趋势，生成机构级行业深度研究报告。
## 触发场景
当用户出现以下任一表述时触发本技能：
- "对 XX 行业做深度分析"
- "分析 XX 行业"
- "XX 行业深度研究"
- "研究 XX 行业的投资价值"
- "XX 行业全面分析"industry-one-page-analysis
When your agent uses it
行业一页纸分析技能。当用户要求对某个行业进行简洁分析、快速点评、一页纸分析、速评时使用，包括"XX 行业一页纸"、"快速分析 XX 行业"、"XX 行业速评"、"XX 行业简单分析"、"XX 行业核心要点"等表述。本技能从行业概况、市场规模、竞争格局、行业趋势、投资机会、主要风险六大维度，利用 gildata-aidata 专业金融数据服务获取权威数据，生成简洁的一页纸 Markdown 分析报告，聚焦核心信息，适合快速决策参考。**报告生成后直接在对话中显示完整内容，无需发送文件**。
What it loads into context
Frontmatter
167 tokensAlways available to the agent
SKILL.md
4,103 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能定位
- 触发场景
- 数据获取策略
- 分析框架（一页纸六大维度）
- 报告输出格式
- 执行流程
- 注意事项
- 特殊情况处理
- 示例调用
- 与行业深度分析技能对比
- 四个分析技能对比总览
- 技能更新日志
Files
SKILL.mdcore14,289 bytes
Read SKILL.md
# 行业一页纸分析技能
## 技能定位
本技能面向**投资经理、研究员、基金经理**，提供行业快速分析工具。在一页纸篇幅内，聚焦行业最核心的投资信息，帮助快速判断行业投资价值和风险，适合晨会汇报、快速决策、赛道初筛等场景。
与 `industry-deep-analysis` 的区别：
- **深度分析**：八大维度，全面深入，适合深度研究和首次覆盖
- **一页纸分析**：六大核心维度，精炼简洁，适合快速决策和赛道跟踪
## 触发场景
当用户出现以下任一表述时触发本技能：
- "XX 行业一页纸"institutional-research-outline
When your agent uses it
机构调研大纲生成技能。当用户需要为上市公司调研准备提纲、生成调研问题清单、准备机构调研材料时使用，包括"生成 XX 公司调研大纲"、"准备去 XX 公司调研"、"调研 XX 公司前需要问什么问题"、"帮我写一份调研提纲"等表述。本技能针对上市公司首次调研场景，生成标准版深度（1-2 小时准备）的结构化调研大纲，包含公司基本面分析、同业对比、历史调研问题回顾和约 10 个核心调研问题，覆盖战略、业务、财务、风险等维度。**报告生成后直接在对话中显示完整内容，无需发送文件**。
What it loads into context
Frontmatter
178 tokensAlways available to the agent
SKILL.md
6,895 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能定位
- 触发场景
- 执行流程
- 报告结构模板
- 特殊场景处理
- 工具使用说明
- gildata-aidata 工具调用指南
- 数据来源优先级
- 质量检查清单
- 注意事项
- 技能更新日志
Files
SKILL.mdcore23,323 bytesSKILL.mdcore23,323 bytesSKILL.mdcore23,323 bytesSKILL.mdcore23,323 bytesSKILL.mdcore23,323 bytesSKILL.mdcore23,323 bytes
Read SKILL.md
# 机构调研大纲生成技能
## 技能定位
本技能用于**生成上市公司机构调研前的结构化调研大纲**，帮助研究员系统性地准备调研问题，覆盖财务、业务、战略、风险等关键维度。**以 gildata-aidata 金融数据服务为核心数据源**，确保财务数据、估值数据、同业对比的专业性和准确性。
**核心优势**：
- ✅ **gildata-aidata 专业数据源**：所有财务数据、估值数据、一致预期优先通过 gildata-aidata 服务获取
- ✅ **分析框架系统**：覆盖基本面、同业对比、历史问题、核心问题四大模块
- ✅ **问题设计专业**：10-12 个针对性调研问题，避免公开信息重复
- ✅ **输出结构化**：标准化 Markdown 格式报告，直接在对话中展示完整内容
**输出格式**：
- 默认生成结构化 Markdown 格式报告When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
3,868 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能概述
- 核心分析维度
- 执行流程
- 数据获取指引
- 数据处理原则
- 注意事项
Files
SKILL.mdcore13,534 bytes
Read SKILL.md
# 城投主体债务排雷分析技能
## 技能概述
本技能面向证券固定收益研究员，提供城投平台（地方政府融资平台）债务风险的自动化排雷分析能力。通过多维度数据交叉验证，识别潜在信用风险，生成结构化风险报告。
## 核心分析维度
### 1. 区域财政实力评估
- 一般公共预算收入及增速
- 政府性基金收入（土地出让金）
- 财政自给率（一般公共预算收入/一般公共预算支出）
- 上级转移支付依赖度
### 2. 债务压力分析macro-asset-allocation
When your agent uses it
宏观大类资产配置建议生成技能。当用户要求生成资产配置报告、大类资产配置建议、投资组合配置、股债商品配置策略时使用。包括"生成资产配置建议"、"当前适合配置什么资产"、"股债商品怎么配"、"宏观视角资产配置"、"投资组合配置建议"等表述。本技能结合最新宏观经济数据、政策预判、市场环境，运用美林时钟/宏观象限等分析框架，自动生成股票、债券、商品、外汇等大类资产配置建议，标注配置逻辑、权重建议与风险点，输出面向专业投资机构的 Markdown 格式资产配置报告。
What it loads into context
Frontmatter
165 tokensAlways available to the agent
SKILL.md
2,815 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 数据获取策略
- 执行流程
- 注意事项
- 示例调用
- 输出要求
- 技能更新日志
Files
SKILL.mdcore9,784 bytes
Read SKILL.md
# 宏观大类资产配置技能
## 概述
本技能模拟宏观策略研究员的大类资产配置视角，结合最新宏观经济数据、政策预判与市场环境，运用**美林时钟**或**宏观象限**分析框架，自动生成股票、债券、商品、外汇及现金等大类资产的配置建议。报告旨在为专业投资机构提供具有逻辑支撑的配置方向、权重建议与风险提示。
## 数据获取策略
### 核心数据源：gildata-aidata
所有宏观经济数据、资金流向、指数行情及市场观点均优先通过 `gildata-aidata` 服务获取。
| 数据需求 | 推荐工具 | 说明 |
|---------|---------|------|macro-daily-briefing
When your agent uses it
证券宏观研究员日报生成技能。当用户要求生成宏观日报、宏观经济简报、每日宏观速评时使用。本技能自动抓取最新宏观数据、政策动态、市场舆情，生成包含核心观点、数据跟踪、政策解读、资产表现及策略展望的标准化 Markdown 宏观日报。
What it loads into context
Frontmatter
87 tokensAlways available to the agent
SKILL.md
3,574 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 数据获取策略
- 执行流程
- 注意事项
- 示例调用
- 技能更新日志
Files
SKILL.mdcore12,503 bytes
Read SKILL.md
# 证券宏观研究员日报生成技能
## 概述
本技能旨在模拟专业证券宏观研究员的日常工作流，通过自动化抓取和分析最新的宏观经济数据、政策动态、市场舆情及大类资产表现，生成一份结构化、逻辑严密、观点鲜明的**宏观日报（Macro Daily Briefing）**。报告风格贴近卖方宏观首席分析师，注重“边际变化”、“预期差”及“投资启示”。
## 数据获取策略
### 核心数据源：gildata-aidata
所有国内宏观数据、政策、舆情、资金面及行情数据均优先通过 `gildata-aidata` 服务获取。
| 数据需求 | 推荐工具 | 说明 |
|---------|---------|------|macro-data-brief
When your agent uses it
宏观经济数据解读与市场分析技能。当用户要求生成宏观数据解读、宏观经济简报、当日经济数据点评、宏观数据对股债影响分析、经济数据速评时使用。包括"今日宏观数据解读"、"XX月经济数据点评"、"宏观数据对股市影响"、"生成宏观数据简报"、"最新PMI/CPI/社融数据解读"等表述。本技能自动搜索当日或最新发布的中国核心宏观经济数据（PMI、CPI、PPI、社融、M2、进出口、工业增加值、固定资产投资、社零等），对比市场预期值，解读数据含义与预期偏差，分析对货币政策和财政政策的传导逻辑，评估对股市、债市的即时影响，最终生成结构化的Markdown格式宏观数据解读简报。
What it loads into context
Frontmatter
217 tokensAlways available to the agent
SKILL.md
3,198 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 数据获取策略
- 执行流程
- 注意事项
- 输出要求
- 示例调用
- 技能更新日志
Files
SKILL.mdcore10,979 bytes
Read SKILL.md
# 宏观经济数据解读简报生成技能
## 概述
本技能用于自动抓取中国核心宏观经济数据，进行预期对比、政策解读和市场影响分析，生成面向投资专业人士的宏观数据解读简报。
## 数据获取策略
### 核心数据源
**Gildata-AIData (gildata-aidata)** 作为主要数据源，提供全方位的宏观经济数据与投研分析工具：
| 数据需求 | 推荐工具 | 说明 |
|---------|---------|------|macro-policy-analysis
When your agent uses it
宏观政策跟踪与推演分析技能。当用户要求分析最新政策、解读央行货币政策、财政政策影响、推演政策对股债市场影响、跟踪政策落地效果时使用。包括"今日政策解读"、"政策对市场的影响"、"后续政策推演"、"政策跟踪分析"、"宏观政策简报"、"美联储政策影响"、"监管新政解读"等表述。本技能自动抓取全球主要经济体政策动态（美联储、欧央行、海外通胀与大宗商品）、国内政策动态（央行货币政策、财政政策、监管新政）、核心宏观舆情（经济会议、官员讲话、行业政策导向），进行政策意图解读、实施路径分析、中长期影响评估，推演后续政策可能性及节奏，跟踪前期政策落地效果，最终生成结构化的Markdown格式政策推演分析报告。
What it loads into context
Frontmatter
227 tokensAlways available to the agent
SKILL.md
3,505 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 数据获取策略
- 执行流程
- 注意事项
- 输出要求
- 示例调用
- 技能更新日志
Files
SKILL.mdcore12,553 bytes
Read SKILL.md
# 宏观政策跟踪与推演分析技能
## 概述
本技能用于自动抓取全球及中国核心政策动态，进行政策意图解读、预期推演和落地效果跟踪，生成面向投资专业人士的政策分析简报。
## 数据获取策略
### 核心数据源
**Gildata-AIData (gildata-aidata)** 作为主要数据源，提供全方位的政策、会议、讲话及舆情数据：
| 数据需求 | 推荐工具 | 说明 |
|---------|---------|------|macro-risk-monitor
When your agent uses it
宏观风险预警监控技能。当用户要求监控宏观经济数据、预警经济下行/通胀/通缩风险、提示政策转向信号、生成宏观预警报告时使用。本技能实时抓取核心宏观指标，通过对比预期与历史数据识别异动，研判经济衰退、滞胀、通缩或政策收紧风险，并输出包含风险等级、传导逻辑、资产影响及应对建议的结构化 Markdown 预警报告。
What it loads into context
Frontmatter
123 tokensAlways available to the agent
SKILL.md
3,418 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 数据获取策略
- 执行流程
- 注意事项
- 示例调用
- 输出要求
- 技能更新日志
Files
SKILL.mdcore11,733 bytes
Read SKILL.md
# 宏观风险预警监控技能
## 概述
本技能模拟宏观策略研究员的风险监控视角，通过持续跟踪核心宏观经济指标，自动识别数据异动（如大幅低于/高于预期、趋势性拐点），精准预警经济下行、通胀抬头、通缩螺旋及政策转向等关键风险，并输出专业的**宏观经济风险预警报告**。
## 数据获取策略
### 核心数据源：gildata-aidata
所有国内宏观数据、政策、舆情、资金面及行情数据均优先通过 `gildata-aidata` 服务获取。
| 数据需求 | 推荐工具 | 说明 |
|---------|---------|------|macro-weekly-briefing
When your agent uses it
证券宏观研究员周报生成技能。当用户要求生成宏观周报、每周宏观经济回顾、周度宏观策略报告时使用。本技能自动抓取本周宏观数据、政策动态、市场舆情及大类资产表现，生成包含核心观点、数据回顾、政策梳理、资产复盘及下周展望的标准化 Markdown 宏观周报。
What it loads into context
Frontmatter
102 tokensAlways available to the agent
SKILL.md
3,897 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 数据获取策略
- 执行流程
- 注意事项
- 示例调用
- 技能更新日志
Files
SKILL.mdcore13,618 bytes
Read SKILL.md
# 证券宏观研究员周报生成技能
## 概述
本技能旨在模拟专业证券宏观研究员的周报撰写工作流，通过对**一周内**宏观经济数据、政策动态、市场舆情及大类资产表现的系统性复盘，生成一份结构严谨、逻辑连贯、观点鲜明的**宏观周报（Macro Weekly Briefing）**。
与日报不同，周报更侧重于**“趋势的确认”**、**“逻辑的连贯性”**以及**“下周的前瞻性”**。报告风格贴近卖方宏观首席分析师，注重将周度高频数据串联成宏观叙事，并给出明确的投资策略建议。
## 数据获取策略
### 核心数据源：gildata-aidata
所有国内宏观数据、政策、舆情、资金面及行情数据均优先通过 `gildata-aidata` 服务获取。
market-minimal-review
When your agent uses it
证券金融工程研究员市场极简复盘技能。自动抓取行情、资金、波动率数据，生成标准化晨会简报与异动提醒，适用于每日盘后快速复盘及晨会汇报准备。
What it loads into context
Frontmatter
61 tokensAlways available to the agent
SKILL.md
2,667 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 核心分析框架
- 数据采集指引 (gildata-aidata 工具映射)
- 执行流程
- 报告模板 (Markdown)
- 注意事项
Files
SKILL.mdcore8,630 bytes
Read SKILL.md
# 市场极简复盘与异动提醒 (Market Minimal Review & Alerts)
## 概述
本技能专为**证券金融工程研究员**设计，旨在提供**数据驱动的极简市场复盘**。通过自动抓取核心行情、资金流向、波动率及量化异动数据，生成标准化的晨会简报素材与盘后异动提醒，帮助研究员快速掌握市场微观结构与量化信号，辅助投资决策。
**所有数据查询必须使用 `gildata-aidata` 服务**。
## 核心分析框架
### 1. 极简行情复盘 (Market Recap)
* **宽基指数**: 上证、深成、创业板、科创 50、北证 50 涨跌幅及成交额变化。
* **风格指数**: 大盘/小盘 (沪深 300 vs 中证 1000)、成长/价值表现对比。
* **市场广度**: 涨跌家数比、涨停/跌停家数、连板高度 (市场情绪温度计)。market-sentiment-monitor
When your agent uses it
市场情绪与风险预警技能。实时跟踪北向资金、两融余额、龙虎榜资金流向及波动率指标，生成情绪与风险预警信号，提示异常变动，适用于盘中监控与盘后复盘。当用户要求生成市场情绪日报、资金面监控报告、风险预警简报、龙虎榜分析、北向资金点评、两融数据分析、市场情绪体温报告时使用，包括"市场情绪监控"、"今日资金面怎么样"、"北向资金流向"、"两融数据分析"、"龙虎榜解读"、"市场风险预警"、"情绪日报"、"盘中监控"、"盘后复盘"等表述。即使用户只说"看看今天情绪"、"资金面怎么样"、"有没有风险信号"等简短表述也要触发。本技能自动通过 gildata-aidata 专业金融数据服务获取核心情绪与资金数据，以md格式生成最终报告。
What it loads into context
Frontmatter
239 tokensAlways available to the agent
SKILL.md
2,898 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 核心监控维度
- 数据采集指引 (gildata-aidata 工具映射)
- 执行流程
- 报告模板 (Markdown)
- 注意事项
Files
SKILL.mdcore9,513 bytes
Read SKILL.md
# 市场情绪与风险预警监控 (Market Sentiment & Risk Monitor)
## 概述
本技能专注于**微观资金流向**与**市场情绪监控**，旨在为交易者和风控人员提供实时的市场体温计和风险雷达。通过对北向资金、两融杠杆、龙虎榜博弈、波动率等核心数据的量化分析，识别资金异常流动，提前预警潜在的流动性风险或情绪反转。
**所有数据查询必须使用 `gildata-aidata` 服务**。
## 核心监控维度
### 1. 北向资金 (Northbound Capital) - "聪明钱"
* **监控指标**: 沪股通/深股通净流入、累计净流入、前十大成交活跃股。
* **预警信号**:
* 🔴 **外资大幅流出**: 单日净流出 > 50 亿元，或连续 3 日净流出。market-trend-style-analysis
When your agent uses it
证券策略市场大势与风格研判技能。适用于策略研究员进行市场方向判断和风格配置。基于量价、资金、情绪及宏观数据，输出大势研判（牛/熊/震荡）、风格强弱排序（成长/价值/大盘/小盘）及仓位建议，生成标准化 Markdown 策略报告。当用户要求分析市场大势、判断市场方向、进行风格配置、评估仓位时使用，包括"市场大势分析"、"当前市场怎么样"、"风格怎么配"、"现在该买大盘还是小盘"、"成长和价值哪个强"、"市场趋势研判"、"仓位建议"等表述。即使用户只说"看看大盘"、"市场怎么样"、"现在什么风格"等简短表述也要触发。本技能自动通过 gildata-aidata 专业金融数据服务获取核心数据，进行量价与逻辑推演，生成标准化策略研判报告。
What it loads into context
Frontmatter
237 tokensAlways available to the agent
SKILL.md
3,509 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 核心分析框架
- 数据采集指引 (gildata-aidata 工具映射)
- 执行流程
- 报告模板 (Markdown)
- 注意事项
Files
SKILL.mdcore12,122 bytes
Read SKILL.md
# 市场大势与风格研判 (Market Trend & Style Analysis)
## 概述
本技能专为**证券策略研究员**设计，旨在提供系统化的**市场大势研判**与**风格配置建议**。通过综合分析量价趋势、流动性环境、市场情绪及宏观因子，量化评估市场状态，输出明确的指数方向判断、风格强弱排序及动态仓位建议，辅助投资决策。
**所有数据查询必须使用 `gildata-aidata` 服务**。
## 核心分析框架
### 1. 大势研判模型 (Trend & Direction)
* **量价趋势**: 均线系统（多头/空头排列）、关键支撑/阻力位、量价配合（放量上涨/缩量下跌）。
* **流动性**: 宏观流动性（M1/M2 剪刀差、社融增速）、微观流动性（北向资金、ETF 申赎、两融余额）。
* **情绪周期**: 换手率、新增开户数、股票型公募基金发行量、期权隐含波动率 (IV)。policy-flash-briefing
When your agent uses it
证券策略研究员政策快评技能。当发生重大政策发布、突发监管消息、宏观数据超预期、盘中市场剧烈异动时触发。本技能自动抓取事件核心要素，进行即时解读，生成策略应对建议与受影响板块梳理，强调“预期差”与“交易指导”。
What it loads into context
Frontmatter
88 tokensAlways available to the agent
SKILL.md
2,411 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 适用场景
- 数据采集指引 (gildata-aidata 工具映射)
- 执行流程
- 标准报告模板 (Markdown)
- 注意事项
Files
SKILL.mdcore8,370 bytes
Read SKILL.md
# 证券策略研究员政策/事件快评 (Policy & Event Flash Briefing)
## 概述
本技能专为**“交易型策略研究”**设计，旨在对突发政策、重大宏观数据、盘中剧烈异动进行**分钟级/小时级**的快速响应。与深度研报不同，本技能聚焦于**“市场即时反应”**与**“预期差交易”**，为交易员提供直接的决策辅助。
核心逻辑：
1. **定性**：是什么性质的政策/事件？（货币、财政、监管、产业？）
2. **定量/预期**：力度如何？是否超预期？（符合预期=利好兑现，超预期=新主线）
3. **传导**：利好谁？利空谁？情绪影响还是实质影响？
4. **应对**：怎么操作？（追高、低吸、观望、对冲？）
**所有数据查询必须使用 `gildata-aidata` 服务**。
When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
2,534 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定位
- 核心工作流
- 报告模板
- 数据获取指引
- 注意事项
- 输出要求
Files
SKILL.mdcore8,854 bytes
Read SKILL.md
# 金融工程日报生成技能
## 角色定位
你是一名资深证券金融工程研究员，负责每日输出标准化金融工程日报。报告面向量化投资团队，聚焦**数据驱动**的客观分析，涵盖市场概况、因子表现、策略信号、资金情绪四大核心模块。
## 核心工作流
### 第一步：获取基础信息
1. 调用 `finx time` 获取当前日期，确定报告日期
2. 识别交易日特征（月初/月末、季末、重要数据发布日等）
### 第二步：数据采集
quant-factor-tracker
When your agent uses it
证券金融工程研究员因子跟踪与评估技能。当用户要求跟踪量化因子表现、更新因子库、计算因子IC/IR、分析分组超额收益、评估因子有效性、监控因子衰减或生成因子跟踪报告时触发。包括“因子跟踪”、“计算因子IC”、“因子失效分析”、“生成因子评估报告”、“因子库更新”、“看看这些因子最近表现怎么样”、“跑一下IC和分组收益”等表述。本技能自动获取或接收因子截面数据与下期收益率，计算Rank IC、IR、分组超额收益，执行单调性与衰减检验，标记因子强弱状态与失效预警，输出标准化Markdown格式因子跟踪报告。
What it loads into context
Frontmatter
191 tokensAlways available to the agent
SKILL.md
2,567 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 核心分析框架
- 数据采集指引 (gildata-aidata 工具映射)
- 执行流程
- 报告模板 (Markdown)
- 注意事项
Files
SKILL.mdcore8,018 bytes
Read SKILL.md
# 量化因子跟踪与评估 (Quant Factor Tracker)
## 概述
本技能专为**证券金融工程研究员**设计，旨在提供系统化的**量化因子跟踪与有效性评估**。通过自动获取或接收截面数据，计算因子 IC/IR、分组超额收益，执行单调性与衰减检验，标记因子强弱状态与失效预警，生成标准化的 Markdown 格式因子跟踪报告，辅助多因子模型构建与策略迭代。
**所有数据查询必须使用 `gildata-aidata` 服务**。
## 核心分析框架
### 1. 因子有效性评估 (Factor Effectiveness)
* **Rank IC (Information Coefficient)**: 因子值与下期收益率的秩相关系数。
* **IC Mean**: 衡量因子预测能力的强弱。通常 |IC| > 0.03 为有效。
* **ICIR (IC / std(IC))**: 衡量因子预测能力的稳定性。通常 ICIR > 0.5 为优秀。quant-strategy-quick-backtest
When your agent uses it
证券金融工程研究员轻量化策略速回测技能。当用户要求进行策略回测、量化策略测试、简易策略验证、快速计算年化收益/最大回撤/夏普比率、策略过拟合检测时使用。包括"回测一下这个策略"、"测试均线交叉策略"、"策略速评"、"快速回测"、"算一下策略收益"、"策略有效性验证"、"帮我回测下"、"看看这个策略怎么样"、"策略表现如何"、"量化回测"、"策略回测报告"等表述。即使用户只说"回测"、"测试策略"、"算下收益"等简短表述也要触发。本技能自动解析策略规则、获取行情数据或模拟数据、执行回测计算、输出核心绩效指标，并附带过拟合风险提示，生成标准化Markdown格式回测报告。
What it loads into context
Frontmatter
220 tokensAlways available to the agent
SKILL.md
3,710 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 技能定位
- 数据采集指引 (gildata-aidata 工具映射)
- 执行流程
- 常见策略模板
- 注意事项
Files
SKILL.mdcore12,082 bytesSKILL.mdcore12,082 bytes
Read SKILL.md
# 量化策略速回测技能 (Quant Strategy Quick Backtest)
## 技能定位
为金融工程研究员提供**轻量化、快速响应**的策略回测能力，支持简易策略规则的解析与验证，快速输出核心绩效指标，并系统性提示过拟合风险。
> **适用场景**：策略灵感快速验证、晨会策略速评、投研交流中的即时测算、策略框架初步筛选。
> **不适用场景**：高频策略回测、复杂多因子模型、需要精细交易成本模拟的实盘级回测。
## 数据采集指引 (gildata-aidata 工具映射)
**所有行情数据必须通过 `gildata-aidata` 服务获取**，具体工具调用如下：
### 1. 指数行情数据 (作为基准或标的)When your agent uses it
What it loads into context
Frontmatter
0 tokensAlways available to the agent
SKILL.md
5,125 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定位
- 核心工作流
- 报告模板
- 数据获取指引
- 注意事项
- 输出要求
Files
SKILL.mdcore18,838 bytes
Read SKILL.md
# 金融工程周报生成技能
## 角色定位
你是一名资深证券金融工程研究员，负责每周输出标准化金融工程周报。报告面向量化投资团队，聚焦**数据驱动**的客观分析，涵盖市场概况、风格因子、行业因子、情绪信号、策略表现五大核心模块，侧重周度趋势确认与下周前瞻性研判。
## 核心工作流
### 第一步：获取基础信息
1. 调用 `finx time` 获取当前日期，确定报告覆盖周期（通常为周一至周五或最近五个交易日）
2. 识别周度交易特征（月初/月末、季末、重要数据发布周等）
### 第二步：数据采集
sector-allocation
When your agent uses it
证券策略行业比较与配置推荐技能。适用于策略研究员进行行业轮动分析与资产配置决策。当用户要求生成行业配置建议、行业比较分析、行业轮动策略、行业投资价值排序时使用，包括"行业配置分析"、"行业比较"、"行业轮动"、"推荐配置哪些行业"、"行业投资价值排序"、"行业配置建议"、"行业超配低配"、"当前应该买什么行业"等表述。即使用户只说"行业怎么配"、"看看哪些行业值得投"等简短表述也要触发。本技能自动通过 gildata-aidata 专业金融数据服务获取行业估值、资金流向、景气度等核心数据，运用4D多维比较框架（景气、估值、资金、政策催化），量化评估各行业投资价值，输出标准化Markdown格式行业配置排序与深度逻辑推演报告。
What it loads into context
Frontmatter
217 tokensAlways available to the agent
SKILL.md
2,636 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 核心分析框架 (4D 模型)
- 数据采集指引
- 执行流程
- 报告模板 (Markdown)
- 注意事项
Files
SKILL.mdcore9,056 bytes
Read SKILL.md
# 行业比较与配置推荐 (Sector Allocation & Comparison)
## 概述
本技能辅助证券策略研究员进行**行业轮动分析**与**资产配置决策**。通过 gildata-aidata 专业金融数据服务自动化采集行业核心数据，运用多维比较框架（景气、估值、资金、政策），量化评估各行业投资价值，输出标准化的行业配置排序与深度逻辑推演。
## 核心分析框架 (4D 模型)
### 1. 景气维度 (Fundamentals / Momentum) - 40% 权重
* **盈利预测**: 分析师盈利预测上调比例（EPS 上调家数/总家数）。
* **业绩增速**: 营收/净利润同比增速、环比改善情况。
* **高频数据**: 行业高频指标（如汽车销量、水泥价格、半导体出货量、运价指数等）的趋势。
* **景气评分**: 趋势向上且超预期者为高分。
strategy-daily-briefing
When your agent uses it
证券策略研究员策略日报生成技能。当用户要求生成策略日报、股市复盘、市场情绪分析、资金流向汇总时使用。本技能自动抓取当日大盘、风格、行业、资金、情绪数据，进行量价与逻辑推演，生成标准化 Markdown 格式策略日报。
What it loads into context
Frontmatter
83 tokensAlways available to the agent
SKILL.md
3,494 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 数据获取策略
- 执行流程
- 注意事项
- 示例调用
- 技能更新日志
Files
SKILL.mdcore11,780 bytes
Read SKILL.md
# 证券策略研究员策略日报生成技能
## 概述
本技能旨在模拟专业证券策略分析师（Strategist）的日度复盘工作流，通过对**当日**市场指数、风格因子、行业板块、资金流向及微观情绪数据的系统性梳理，生成一份结构严谨、逻辑清晰的**策略日报（Strategy Daily Briefing）**。
策略日报的核心不仅是数据的罗列，更是**“资金行为的解码”**与**“交易主线的确认”**。报告风格贴近买方/卖方策略团队，注重从微观结构中洞察市场拐点，并给出明确的下个交易日交易计划。
## 数据获取策略
### 核心数据源：gildata-aidata
所有行情数据、资金流向及市场情绪数据均优先通过 `gildata-aidata` 服务获取。
valuation-prosperity-tracking
When your agent uses it
证券策略行业估值与景气度跟踪技能。适用于策略研究员进行行业比较和轮动决策。自动获取并分析行业估值分位、高频景气数据，对比历史数据，预警估值泡沫与景气拐点，生成标准化 Markdown 跟踪报告。
What it loads into context
Frontmatter
82 tokensAlways available to the agent
SKILL.md
2,724 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 概述
- 核心分析框架
- 数据采集指引 (gildata-aidata 工具映射)
- 执行流程
- 报告模板 (Markdown)
- 注意事项
Files
SKILL.mdcore8,621 bytes
Read SKILL.md
# 行业估值与景气跟踪 (Valuation & Prosperity Tracking)
## 概述
本技能专为**证券策略研究员**设计，旨在提供系统化的**行业估值分位监控**与**高频景气度跟踪**。通过实时采集与分析多维数据，识别行业性价比，预警估值泡沫与景气拐点，辅助行业配置与轮动决策。
**所有数据查询必须使用 `gildata-aidata` 服务**。
## 核心分析框架
### 1. 估值跟踪模型 (Valuation Monitoring)
* **绝对估值**: PE (TTM)、PB (MRQ)。
* **相对估值**: 历史分位数 (1年、3年、5年、10年)，风险溢价 (ERP)。
* **泡沫预警**:insurance-underwriting-conclusion-interpretation
When your agent uses it
将专业核保结论转化为客户可理解的通俗话术，并生成标准化的核保结果通知书。内置合规过滤与易混淆点预警，确保客户准确理解核保结论。覆盖通俗话术生成、易混淆点预警、标准体承保通知、加费/除外承保通知、延期核保通知、拒保通知等全流程。当核保评估完成需要向客户解释结论、出具正式核保结论、生成加费/除外条款说明书、通知被保险人补充材料或告知拒保原因时触发。适用于核保结论通俗解读、核保结案通知生成、拒保函件起草、加费/除外承保条件确认书制作场景。 触发词：核保结论、结论解释、结论解读、除外说明、通俗话术。
What it loads into context
Frontmatter
265 tokensAlways available to the agent
SKILL.md
4,481 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 并行技能输出合并规则
- 工作流程
- 输出格式
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset257 bytesSKILL.mdcore15,676 bytesconfig.jsonasset1,019 bytesrequirements.txtdoc173 bytesscripts/validate_input.pyruns1,971 bytes
Read SKILL.md
# 核保结论解读与通知书生成
## 角色定义
扮演核保结论解读与通知书起草专家。将专业核保结论转化为客户可理解的通俗话术，识别易混淆点并预警，同时生成符合监管要求、语言规范的核保结果通知书，确保客户准确理解结论含义，通知内容完整、依据充分、表述合规。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 核保结论解读
- 核保结论解释
- 通俗话术
- 核保结果通知insurance-underwriting-follow-up-outreach
When your agent uses it
主动联系客户进行回访，支持电话拨打与消息发送，基于客户反馈识别真实意图（确认/咨询/投诉/退保/加保），输出结构化标签与跟进建议。当需要回访客户、拨打回访电话、发送回访消息、识别客户意图时触发。适用于保险核保后送达回访、客户意图识别、客户关怀跟进场景。 触发词：回访客户、打电话、发消息、客户意图。
What it loads into context
Frontmatter
163 tokensAlways available to the agent
SKILL.md
2,373 tokensWhen the skill is activated
Reference documents
465 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 系统依赖
- MCP工具调用
- 输出格式
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset123 bytesSKILL.mdcore8,435 bytesTEST.mddoc1,595 bytesconfig.jsonasset445 bytesrequirements.txtdoc145 bytesscripts/validate_input.pyruns1,881 bytes
Read SKILL.md
# 回访跟进
## 角色定义
扮演保险回访跟进专家。主动联系客户进行回访，支持电话拨打与消息发送，基于客户反馈识别真实意图（确认/咨询/投诉/退保/加保），输出结构化标签与跟进建议，确保客户权益得到保障并挖掘潜在服务机会。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 回访客户
- 打电话
- 发消息
- 客户意图insurance-underwriting-health-verification
When your agent uses it
对投保申请进行全面核保健康审核，涵盖客户自述语音解析、术语标准化、健康告知逐项核对与疾病核保规则匹配。对健康告知及客户自述语音进行语义解析、术语标准化与交叉矛盾校验，基于多维度风险评分（健康/财务/家庭）、风险因素加权评估，输出综合风险等级与核保结论建议。当需要审核投保申请、评估投保风险、判断是否可承保、核对健康告知、评估客户风险等级时触发。适用于核保人员新单审核、代理人投保预检、健康告知评估、客户风险画像生成场景。 触发词：健康告知、客户自述、告知矛盾、术语标准化。
What it loads into context
Frontmatter
235 tokensAlways available to the agent
SKILL.md
8,730 tokensWhen the skill is activated
Reference documents
2,260 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 输出格式
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset116 bytesSKILL.mdcore28,689 bytesconfig.jsonasset1,046 bytesreferences/output-template.mddoc2,679 bytesreferences/underwriting-rules.mddoc3,599 bytesrequirements.txtdoc146 bytesscripts/validate_input.pyruns2,056 bytes
Read SKILL.md
# 投保健康审核
## 角色定义
扮演保险核保专家。对投保申请进行全面的核保健康审核，融合客户多维度风险评估、投保材料深度结构化解析、健康告知逐项核对与风险因素加权评估，标记风险等级并给出核保结论建议及保障配置方向。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 核保预审
- 投保审核
- 投保风险
- 健康告知insurance-underwriting-medical-assessment
When your agent uses it
对被保险人医疗材料进行全面核保医学评估，涵盖体检报告深度解析、病历与病理材料结构化提取、异常指标核保风险判定、多疾病交叉推理及核保结论输出。当需要解读体检报告以支持核保决策、评估体检异常指标的核保影响、解析病历/病理材料、判断是否需要要求加做体检项目、输出含具体加费比例与除外责任描述的核保结论时触发。适用于健康险/寿险核保中的体检报告评估、病历风险分析、加费/除外/拒保依据分析、多疾病交叉核保推理、被保险人医疗结果核保咨询场景。 触发词：体检报告、病历风险、异常指标。
What it loads into context
Frontmatter
240 tokensAlways available to the agent
SKILL.md
5,302 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 输出格式
- 系统依赖
- MCP工具调用
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset257 bytesSKILL.mdcore17,845 bytesconfig.jsonasset1,190 bytesrequirements.txtdoc166 bytesscripts/validate_input.pyruns1,615 bytes
Read SKILL.md
# 体检报告与病历核保医学评估
## 角色定义
扮演核保医学顾问。对被保险人提交的体检报告、病历资料、病理材料等进行深度结构化解析与逐项核保分析，将医疗异常指标转化为核保风险评估结论，支持多疾病交叉推理，输出含具体加费比例与除外责任描述的核保建议，为核保人员提供专业的医学依据和处理建议。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 体检报告核保
- 体检异常核保影响
- 异常指标
- 病历风险insurance-underwriting-recording-inspection
When your agent uses it
对销售双录音视频进行全量转写、话术合规比对、说话人分离与情绪异常标记。当需要质检双录音视频、检查销售话术是否合规、识别双录中的情绪异常时触发。适用于保险销售双录质检、话术合规审核、监管合规管理场景。 触发词：双录质检、双录问题、话术合规。
What it loads into context
Frontmatter
148 tokensAlways available to the agent
SKILL.md
2,354 tokensWhen the skill is activated
Reference documents
450 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 角色定义
- 触发条件
- 前置条件
- 工作流程
- 系统依赖
- MCP工具调用
- 输出格式
- 关联技能
- 合规约束
- 审计日志
- Gotchas（踩坑记录）
- 测试用例
- 结束条件
Scripts executed separately
scripts/validate_input.py
Files
.env.exampleasset123 bytesSKILL.mdcore8,220 bytesTEST.mddoc1,554 bytesconfig.jsonasset514 bytesrequirements.txtdoc147 bytesscripts/validate_input.pyruns1,766 bytes
Read SKILL.md
# 双录质检
## 角色定义
扮演保险双录合规质检专家。对销售双录音视频进行全量转写，逐句比对话术合规性，分离说话人身份，标记情绪异常片段，输出结构化质检报告，确保销售过程符合监管要求。
## 触发条件
当用户提及或需要进行以下场景时触发：
- 双录质检
- 双录问题
- 话术合规
- 双录审核client-kyc-profiling
When your agent uses it
基于客户信息和持仓数据构建深度KYC画像，分析家庭生命周期、 风险承受力和资产配置偏好，为后续经营提供基础。 当用户提到客户画像、KYC分析、了解客户、客户情况分析、家庭分析时触发。 不用于持仓诊断（由portfolio-health-check处理）。
What it loads into context
Frontmatter
92 tokensAlways available to the agent
SKILL.md
1,026 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore3,611 bytes
Read SKILL.md
# 客户KYC画像
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`AnalyzeFamilyMembers`, `AnalyzeFinancialIndicators`, `AnalyzeAssetLiability`, `GetFundAssetClassAnalysis`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据client-opportunity-scan
When your agent uses it
识别高价值商机客户，基于到期产品、资产变动、财富阶梯升级等维度 筛选优先联系客户清单并给出经营建议。 当用户提到找商机、哪些客户要联系、客户筛选、蓄客名单、优先联系谁时触发。 不用于已有持仓的分析诊断（由portfolio-health-check处理）。
What it loads into context
Frontmatter
104 tokensAlways available to the agent
SKILL.md
1,074 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore3,616 bytes
Read SKILL.md
# 客户商机扫描
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能可选工具：`GetCurrentTime`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据maturity-client-alert
When your agent uses it
针对理财产品到期或即将到期的客户，生成承接方案， 包含到期产品分析、替代产品推荐和承接话术。 当用户提到到期承接、产品到期、客户续投、封闭期结束、理财到期时触发。 不用于一般产品推荐（由smart-product-matching处理）。
What it loads into context
Frontmatter
95 tokensAlways available to the agent
SKILL.md
1,391 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore4,828 bytes
Read SKILL.md
# 到期客户承接
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`BatchGetFundsDetail`, `SearchFunds`, `GetBatchFundPerformance`, `BatchGetFundTradeRules`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据client-care-script
When your agent uses it
生成客户关怀话术，覆盖生日祝福、节日问候、亏损安抚、流失挽留场景。 侧重情感维护和客户关系保持。 当用户提到安抚客户、亏损沟通、生日祝福、节日问候、流失挽留、客户情绪不好时触发。 不用于主动营销触达（由outreach-script-generator处理）。
What it loads into context
Frontmatter
108 tokensAlways available to the agent
SKILL.md
1,064 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore3,544 bytes
Read SKILL.md
# 客户关怀话术
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`GetLatestQuotations`, `GetBatchFundPerformance`, `BatchGetFundNavHistory`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据hot-product-briefing
When your agent uses it
将主推产品转化为卖点话术和FAQ应答，包含产品亮点、 适配客群画像、常见异议应对。 当用户提到主推产品、产品介绍、怎么跟客户讲这个产品、卖点提炼、产品话术时触发。 不用于产品筛选推荐（由smart-product-matching处理）。
What it loads into context
Frontmatter
95 tokensAlways available to the agent
SKILL.md
1,134 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore3,898 bytes
Read SKILL.md
# 主推产品卖点
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`BatchGetFundsDetail`, `GetBatchFundPerformance`, `GetFundDiagnosis`, `AnalyzeFundRisk`, `BatchGetFundTradeLimit`, `SearchFunds`, `GuessFundCode`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据market-event-interpreter
When your agent uses it
将市场突发事件翻译成客户听得懂的语言，生成事件简析、 对持仓影响评估和建议行动话术。 当用户提到市场大跌怎么说、政策怎么解读、波动话术、热点事件影响、 突发事件怎么跟客户解释时触发。 不用于常规行情查询（由market-insight-qa处理）， 不用于生成完整早报（由market-morning-brief处理）。
What it loads into context
Frontmatter
117 tokensAlways available to the agent
SKILL.md
1,265 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore4,391 bytes
Read SKILL.md
# 市场事件解读
## 可用工具
本技能可调用以下 MCP 数据服务，根据事件类型智能路由：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`SearchFinancialNews`, `searchRealtimeAiAnalysis`, `GetLatestQuotations`, `SearchManagerViewpoint`, `RenderEchart`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据market-morning-brief
When your agent uses it
自动整合行情数据、热点资讯、AI解读和机构观点，生成专业且易读的每日市场早报， 回答"昨天发生了什么、为什么发生、对我有什么影响"三个核心问题。 当用户提到今天市场怎么样、生成早报、晨会内容、市场快报、昨天行情、每日简报时触发。 不用于回答具体单一行情数据问题（如"沪深300现在多少点"），该场景由market-insight-qa处理。
What it loads into context
Frontmatter
124 tokensAlways available to the agent
SKILL.md
1,836 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 核心原则
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore6,137 bytes
Read SKILL.md
# 晨会市场早报
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`GetCurrentTime`, `GetLatestQuotations`, `SearchFinancialNews`, `SearchHotTopic`, `searchRealtimeAiAnalysis`, `SearchManagerViewpoint`, `RenderEchart`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据outreach-script-generator
When your agent uses it
结合市场行情和客户持仓情况，生成个性化电话触达话术， 覆盖资讯同步、风险提醒、产品到期承接、服务邀约等场景。 当用户提到打电话话术、触达话术、通联话术、电话联系客户、怎么跟客户打电话时触发。 不用于客户安抚挽留场景（由client-care-script处理）。
What it loads into context
Frontmatter
106 tokensAlways available to the agent
SKILL.md
1,374 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore4,774 bytes
Read SKILL.md
# 电话触达话术生成器
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`GetLatestQuotations`, `SearchFinancialNews`, `GetBatchFundPerformance`, `BatchGetFundsDetail`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据wechat-moments-creator
When your agent uses it
为理财经理创作专业且有温度的企微/朋友圈营销内容， 支持市场点评、产品解读、投教科普、节日关怀四种风格。 当用户提到朋友圈文案、企微内容、私域运营、营销内容、发条朋友圈时触发。 不用于电话触达话术（由outreach-script-generator处理）。
What it loads into context
Frontmatter
101 tokensAlways available to the agent
SKILL.md
929 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore3,270 bytes
Read SKILL.md
# 朋友圈内容创作
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`GetLatestQuotations`, `SearchFinancialNews`, `SearchHotTopic`, `BatchGetFundsDetail`, `GetBatchFundPerformance`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据client-segmentation-strategy
When your agent uses it
根据AUM、风险偏好、持仓特征进行客户分群， 匹配差异化经营策略和优先级排序。 当用户提到客户分群、分层经营、客群运营、怎么管理客户、客户分类时触发。 不用于单个客户画像分析（由client-kyc-profiling处理）。
What it loads into context
Frontmatter
92 tokensAlways available to the agent
SKILL.md
1,160 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore3,904 bytes
Read SKILL.md
# 客户分群策略
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能可选工具：`GetFundAssetClassAnalysis`, `AnalyzeFinancialIndicators`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据competitor-product-compare
When your agent uses it
对比同类基金产品，从收益、风险、费率、流动性、基金经理能力五个维度 生成对比报告，帮助客户经理回应"别家产品更好"的异议。 当用户提到跟XX比怎么样、竞品对比、同类比较、为什么选我们、 两只基金比较时触发。 不用于单只基金分析（由fund-deep-research处理）。
What it loads into context
Frontmatter
108 tokensAlways available to the agent
SKILL.md
2,033 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 核心原则
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore6,362 bytes
Read SKILL.md
# 竞品对比分析
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`SearchFunds`, `GuessFundCode`, `BatchGetFundsDetail`, `GetBatchFundPerformance`, `AnalyzeFundRisk`, `BatchGetFundTradeRules`, `RenderEchart`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据sales-sop-navigator
When your agent uses it
基于客户所处投资阶段，提供标准化销售SOP步骤引导， 包括开场白、需求挖掘、方案呈现、异议处理、促成话术的完整流程指导。 当用户提到销售流程、SOP、成交技巧、面谈流程、怎么谈客户、 需求挖掘、异议处理时触发。 不用于生成具体产品话术（由hot-product-briefing处理）。
What it loads into context
Frontmatter
115 tokensAlways available to the agent
SKILL.md
1,014 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore3,390 bytes
Read SKILL.md
# 销售SOP导航
## 可用工具
本技能可调用以下 MCP 数据服务，MCP 工具为可选补充，本技能以销售方法论知识为主：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能可选工具：按需调用（如话术中需引用产品数据时使用）
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据compliance-risk-qa
When your agent uses it
回答销售合规问题，包括适当性管理、风险提示义务、 销售话术合规性审查、禁止行为边界。 当用户提到这样说合规吗、能不能承诺收益、风险提示怎么说、 适当性匹配、代客操作合规吗时触发。 不用于具体产品知识问答（由product-knowledge-qa处理）。
What it loads into context
Frontmatter
103 tokensAlways available to the agent
SKILL.md
812 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore2,867 bytes
Read SKILL.md
# 合规风控问答
## 可用工具
本技能可调用以下 MCP 数据服务，仅在需要查询具体法规或产品合规信息时使用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能可选工具：按需调用（本技能以合规知识问答为主，MCP 工具为辅助补充）
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据market-insight-qa
When your agent uses it
快速回答市场行情类问题，包括指数表现、板块走势、 市场温度、北向资金等，返回简洁精准的数据和简评。 当用户提到大盘怎么样、哪些板块表现好、A股走势、 黄金价格、北向资金时触发。 不用于生成完整早报（由market-morning-brief处理）。
What it loads into context
Frontmatter
94 tokensAlways available to the agent
SKILL.md
1,079 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore3,761 bytes
Read SKILL.md
# 市场行情问答
## 可用工具
本技能可调用以下 MCP 数据服务，根据问题类型智能路由：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`GetLatestQuotations`, `SearchFinancialNews`, `SearchHotTopic`, `searchRealtimeAiAnalysis`, `SearchFunds`, `RenderEchart`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据product-knowledge-qa
When your agent uses it
快速解答产品费率、交易规则、分红规则、申赎限制、业绩基准等 日常产品知识问题，支持基金、理财产品、债券等多品类。 当用户提到费率多少、什么时候到账、分红方式、怎么买、 交易规则、起购金额、能不能定投时触发。 不用于生成完整的产品卖点话术（由hot-product-briefing处理）。
What it loads into context
Frontmatter
115 tokensAlways available to the agent
SKILL.md
947 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore3,463 bytes
Read SKILL.md
# 产品知识问答
## 可用工具
本技能可调用以下 MCP 数据服务，根据产品类型智能路由：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`SearchFunds`, `GuessFundCode`, `BatchGetFundsDetail`, `BatchGetFundTradeRules`, `BatchGetFundTradeLimit`, `BatchGetFundsDividendRecord`, `getFundBenchmarkInfo`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据fund-deep-research
When your agent uses it
对单只基金进行全面尽调分析，按基金类型（股票型/债券型/混合型/指数型）自动选择分析路径， 输出业绩、风险、持仓、基金经理能力四维诊断报告。 当用户提到分析基金、基金怎么样、能不能买、基金诊断、基金尽调、这个基金经理靠谱吗时触发。 不用于多只基金的组合持仓分析（由portfolio-health-check处理）。
What it loads into context
Frontmatter
114 tokensAlways available to the agent
SKILL.md
2,926 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 核心原则
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore9,724 bytes
Read SKILL.md
# 基金深度尽调
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`SearchFunds`, `GuessFundCode`, `BatchGetFundsDetail`, `GetBatchFundPerformance`, `AnalyzeFundRisk`, `GetFundDiagnosis`, `BatchGetFundsHolding`, `getFundIndustryAllocation`, `getFundIndustryConcentration`, `getFundTurnoverRate`, `getFundBrinsonIndicator`, `getMarketTimingIndicator`, `getBondAllocationByFundCode`, `getBondFundCreditRatingLevel`, `getBondIndicator`, `getFundCampisiIndicator`, `RenderEchart`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据portfolio-health-check
When your agent uses it
对客户全部持仓进行穿透式体检，从资产配置、基金相关性、回测表现、风险暴露四个维度 输出诊断报告和可视化图表。 当用户提到持仓分析、组合诊断、持仓体检、配置检视、帮我看看持仓时触发。 不用于单只基金分析（由fund-deep-research处理）， 不用于专项风险排查（由portfolio-risk-radar处理）。
What it loads into context
Frontmatter
117 tokensAlways available to the agent
SKILL.md
2,668 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 核心原则
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore8,629 bytes
Read SKILL.md
# 持仓健康诊断
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`DiagnoseFundPortfolio`, `GetFundAssetClassAnalysis`, `BatchGetFundsDetail`, `GetBatchFundPerformance`, `AnalyzePortfolioRisk`, `GetFundsCorrelation`, `GetFundsBackTest`, `GuessFundCode`, `RenderEchart`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据portfolio-risk-radar
When your agent uses it
对投资组合进行专项风险扫描，识别集中度、相关性、流动性、 风格漂移等风险因子，输出风险预警和对冲建议。 当用户提到风险评估、组合风险、风险预警、集中度过高、 相关性分析、最大回撤时触发。 不用于全面持仓体检（由portfolio-health-check处理）。
What it loads into context
Frontmatter
100 tokensAlways available to the agent
SKILL.md
2,293 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 核心原则
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore7,287 bytes
Read SKILL.md
# 组合风险雷达
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`AnalyzePortfolioRisk`, `AnalyzeFundRisk`, `GetFundsCorrelation`, `GetAssetAllocation`, `fund-equity-position`, `fund-recovery-ability`, `RenderEchart`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据asset-allocation-optimizer
When your agent uses it
基于客户风险偏好和当前持仓，结合市场环境和量化模型， 生成个性化资产配置方案并模拟预期收益分布。 当用户提到资产配置、配置优化、怎么调仓、投资方案设计、配置方案时触发。 不用于单只基金分析（由fund-deep-research处理）， 不用于具体产品推荐（由smart-product-matching处理）。
What it loads into context
Frontmatter
105 tokensAlways available to the agent
SKILL.md
2,307 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 核心原则
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore7,473 bytes
Read SKILL.md
# 资产配置优化
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`GetAssetAllocationPlan`, `GetCompositeModel`, `GetFundAssetClassAnalysis`, `MonteCarloSimulate`, `GetLatestQuotations`, `RenderEchart`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据family-financial-planner
When your agent uses it
家庭理财规划师，模拟 CFP 注册理财规划师的完整工作流。通过结构化对话采集家庭成员、资产负债、收支情况，进行财务健康诊断、现金流推演和资产配置建议，输出专业理财规划方案。当用户提到理财规划、家庭资产配置、财务规划、我的钱怎么安排、未来养老规划、子女教育规划时触发。
What it loads into context
Frontmatter
101 tokensAlways available to the agent
SKILL.md
2,956 tokensWhen the skill is activated
Reference documents
1,780 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 角色定位
- 核心工作流
- 参考文档
Files
SKILL.mdcore10,525 bytesplanning-logic.mddoc5,287 bytes
Read SKILL.md
# 家庭理财规划师（Family Financial Planner）
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`AnalyzeFamilyMembers`, `AnalyzeAssetLiability`, `AnalyzeIncomeExpense`, `AnalyzeFinancialIndicators`, `AnalyzeCashFlow`, `GetAssetAllocationPlan`, `GetCompositeModel`, `MonteCarloSimulate`, `AnalyzeInvestmentPerformance`, `RenderEchart`, `RenderHtmlToPdf`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据investment-simulation
When your agent uses it
运用蒙特卡洛模拟对拟投资组合进行多周期收益概率测算， 可视化乐观/中性/悲观三种情景，帮助客户理解预期风险收益。 当用户提到收益模拟、能赚多少、预期收益、蒙特卡洛、 投资回报、投多少能赚多少时触发。 不用于资产配置方案设计（由asset-allocation-optimizer处理）。
What it loads into context
Frontmatter
107 tokensAlways available to the agent
SKILL.md
2,119 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 核心原则
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore6,406 bytes
Read SKILL.md
# 投资收益模拟
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`MonteCarloSimulate`, `GetFundsBackTest`, `AnalyzePortfolioRisk`, `RenderEchart`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据smart-product-matching
When your agent uses it
从客户配置缺口出发筛选匹配产品，经尽调验证后输出推荐方案和 "为什么适合您"的销售话术。支持基金、理财产品、债券等多品类匹配。 当用户提到推荐产品、该买什么、产品匹配、有什么好产品、推荐基金、理财产品推荐时触发。 不用于已持有产品的诊断分析（由portfolio-health-check处理）， 不用于资产配置方案设计（由asset-allocation-optimizer处理）。
What it loads into context
Frontmatter
128 tokensAlways available to the agent
SKILL.md
2,091 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 核心原则
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore7,050 bytes
Read SKILL.md
# 产品智能匹配
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`SearchFunds`, `GetPopularFund`, `BatchGetFundsDetail`, `GetBatchFundPerformance`, `AnalyzeFundRisk`, `GetFundDiagnosis`, `RenderEchart`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据daily-work-review
When your agent uses it
整合当日工作数据生成复盘报告，包含通联分析、成交总结、 流失原因、改进建议，帮助客户经理日日精进。 当用户提到今天复盘、日报、日终总结、业绩复盘、今天工作怎么样时触发。 不用于单个客户服务记录（由service-summary-generator处理）。
What it loads into context
Frontmatter
90 tokensAlways available to the agent
SKILL.md
845 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore2,887 bytes
Read SKILL.md
# 日终工作复盘
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能可选工具：`GetCurrentTime`, `GetLatestQuotations`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据service-summary-generator
When your agent uses it
根据当日客户服务内容自动生成结构化服务记录，包含服务要点、 客户反馈、后续待办，可直接用于CRM系统录入。 当用户提到写服务记录、服务小结、拜访总结、工单生成、记录沟通内容时触发。 不用于日终整体复盘（由daily-work-review处理）。
What it loads into context
Frontmatter
90 tokensAlways available to the agent
SKILL.md
764 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore2,642 bytes
Read SKILL.md
# 服务记录小结
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能可选工具：`GetCurrentTime`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据tomorrow-plan-generator
When your agent uses it
基于复盘结果和待办事项生成次日工作计划，包含重点客户排序、 优先任务、时间安排建议。 当用户提到明天计划、明日安排、工作规划、下一步做什么、 明天先做什么时触发。 不用于当日工作复盘（由daily-work-review处理）。
What it loads into context
Frontmatter
86 tokensAlways available to the agent
SKILL.md
920 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore2,929 bytes
Read SKILL.md
# 明日工作计划
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能可选工具：`GetCurrentTime`, `GetTxnDayRange`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据investor-education-qa
When your agent uses it
用通俗语言解释金融概念和投资原理，输出可直接发给客户的科普内容。 当用户提到什么是夏普比率、基金知识、投教内容、 帮客户解释XX概念、写段科普时触发。 不用于具体产品问答（由product-knowledge-qa处理）。
What it loads into context
Frontmatter
89 tokensAlways available to the agent
SKILL.md
757 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore2,650 bytes
Read SKILL.md
# 投教知识科普
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`SearchFunds`, `BatchGetFundsDetail`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据market-hotspot-digest
When your agent uses it
将当日市场热点翻译成通俗语言，生成可直接转发给客户的简短解读卡片。 面向"给客户发内容"场景。 当用户提到今天热点、帮我解读热点、给客户发点内容、 市场热点速递、最近值得关注的事时触发。 不用于生成完整早报（由market-morning-brief处理）， 不用于突发事件应对话术（由market-event-interpreter处理）。
What it loads into context
Frontmatter
123 tokensAlways available to the agent
SKILL.md
848 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore3,036 bytes
Read SKILL.md
# 市场热点速递
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`SearchHotTopic`, `SearchFinancialNews`, `searchRealtimeAiAnalysis`, `GetLatestQuotations`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据portfolio-alert-narrator
When your agent uses it
监测持仓基金异动信号（净值大幅波动、分红公告、基金经理变更）， 生成客户可理解的解读和应对建议。 当用户提到基金跌了怎么办、净值异动、基金分红了、 基金经理换了、基金有什么公告时触发。 不用于系统性市场事件解读（由market-event-interpreter处理）。
What it loads into context
Frontmatter
106 tokensAlways available to the agent
SKILL.md
999 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore3,429 bytes
Read SKILL.md
# 持仓异动解读
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`BatchGetFundNavHistory`, `GetFundAnnouncements`, `GetAnnouncementContent`, `BatchGetFundsDetail`, `getBondFundWithAlertRecord`, `getFundDiveCount`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据skill-name-here
When your agent uses it
简要描述这个Skill做什么（1-2句话）。当用户提到XX、YY、ZZ时触发。 不用于XX场景（负触发条件）。
What it loads into context
Frontmatter
50 tokensAlways available to the agent
SKILL.md
525 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 可用工具
- 输入要求
- 执行流程
- 输出模板
- 注意事项
Files
SKILL.mdcore1,844 bytes
Read SKILL.md
# Skill 中文名称
## 可用工具
本技能可调用以下 MCP 数据服务，执行流程中按需选用：
**盈米金融数据（qieman）**
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse`
- 核心能力：基金搜索/诊断、组合分析/回测、资产配置方案、CFP 工具链、图表渲染
- 本技能主要工具：`工具A`, `工具B`
**恒生聚源金融数据（gildata-aidata）**
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
- 核心能力：个股研究(A/H/US)、财务报表、资金流向、研报舆情、理财产品、宏观数据Skills included
- insurance-claim-adjudication-review
- insurance-claim-adjustment
- insurance-claim-case-registration
- insurance-claim-document-processing
- insurance-claim-fraud-detection
- insurance-claim-notification
- insurance-claim-coverage-analysis
- insurance-claim-medical-review
- credit-due-diligence
- credit-industry-analysis
- equity-penetration-analysis
- financial-report-analysis
- pre-visit-plan
- product-roadshow
- submit-credit-application
- visit-memo
- admission-rules-scan
- ai-risk-planning
- credit-case-intake-check
- credit-collateral-risk-mgmt
- credit-related-party-detection
- pre-visit-credit-analysis
- reviewer-visit-memo
- credit-industry-rule-gen
- credit-large-exposure-mgmt
- credit-policy-analysis
- credit-risk-cot
- credit-risk-extraction
- loan-risk-monitor
- post-loan-management
- vlm-verifier
- auto-experiment
- data-profiling
- dnn-modeling
- dnn-tuning
- feature-analysis
- lr-modeling
- lr-tuning
- model-comparison
- model-explanation
- segment-modeling
- univariate-analysis
- xgb-deepmodel
- xgb-modeling
- xgb-tuning
- insurance-agent-activity-planner
- insurance-agent-application-check
- insurance-agent-client-consulting
- insurance-agent-customer-profiling
- insurance-agent-gap-advisor
- insurance-agent-health-disclosure-guide
- insurance-agent-objection-handler
- insurance-agent-plan-generator
- insurance-agent-policy-explanation
- insurance-agent-product-wiki
- insurance-agent-renewal-assist
- insurance-agent-social-media
- a-market-hotspot-discovery
- comparable-company-analysis
- fund-diagnosis
- fund-multi-factor-filter
- stock-fund-analysis
- stock-multi-factor-filter
- stock-quote-analysis
- stock-shareholder-analysis
- stock-tech-analysis
- announcement-analysis
- company-deep-analysis
- company-one-page-analysis
- convertible-bond-valuation
- fixed-income-daily-brief
- fixed-income-weekly-brief
- global-finance-brief
- global-macro-linkage
- industry-deep-analysis
- industry-one-page-analysis
- institutional-research-outline
- macro-asset-allocation
- macro-daily-briefing
- macro-data-brief
- macro-policy-analysis
- macro-risk-monitor
- macro-weekly-briefing
- market-minimal-review
- market-sentiment-monitor
- market-trend-style-analysis
- policy-flash-briefing
- quant-factor-tracker
- quant-strategy-quick-backtest
- sector-allocation
- strategy-daily-briefing
- valuation-prosperity-tracking
- insurance-underwriting-conclusion-interpretation
- insurance-underwriting-follow-up-outreach
- insurance-underwriting-health-verification
- insurance-underwriting-medical-assessment
- insurance-underwriting-recording-inspection
- client-kyc-profiling
- client-opportunity-scan
- maturity-client-alert
- client-care-script
- hot-product-briefing
- market-event-interpreter
- market-morning-brief
- outreach-script-generator
- wechat-moments-creator
- client-segmentation-strategy
- competitor-product-compare
- sales-sop-navigator
- compliance-risk-qa
- market-insight-qa
- product-knowledge-qa
- fund-deep-research
- portfolio-health-check
- portfolio-risk-radar
- asset-allocation-optimizer
- family-financial-planner
- investment-simulation
- smart-product-matching
- daily-work-review
- service-summary-generator
- tomorrow-plan-generator
- investor-education-qa
- market-hotspot-digest
- portfolio-alert-narrator
- skill-name-here
Repository facts
- License
- MIT
- Created
- Apr 21, 2025
- Last pushed
- Oct 17, 2026
- Forks
- 68
- Open issues
- 2
- Project activity
- Active
Rank over the last 12 weeks
Now #982 · best #753 · down 132 in 30 days
- #953 on Oct 8
- #918 on Oct 9
- #928 on Oct 12
- #943 on Oct 15
- #965 on Oct 20
- #982 since Oct 21
28 earlier periods
- #753 on Aug 2
- #762 on Aug 8
- #807 on Aug 10
- #806 on Aug 11
- #796 on Aug 16
- #804 on Aug 17
- #822 on Aug 18
- #819 on Aug 19
- #815 on Aug 21
- #822 on Aug 23
- #835 on Aug 24
- #830 on Aug 26
- #843 on Aug 27
- #877 on Aug 28
- #807 on Aug 30
- #852 on Sep 2
- #856 on Sep 3
- #821 on Sep 4
- #877 on Sep 8
- #833 on Sep 13
- #866 on Sep 14
- #874 on Sep 16
- #850 on Sep 20
- #892 on Sep 22
- #881 on Sep 23
- #845 on Sep 24
- #890 on Sep 28
- #919 on Sep 29
Before you install
Local scriptsAutomated signals, not a guarantee
Checked Nov 5, 2026
Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.
Show 125 notes
DianJin-SKILLS/claim-expert/claim-case-registration/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/claim-expert/claim-document-processing/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/claim-expert/claim-document-processing/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/claim-expert/claim-document-processing/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/claim-expert/claim-document-processing/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/claim-expert/claim-document-processing/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/claim-expert/claim-document-processing/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/claim-expert/claim-document-processing/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/corporate-banker/submit-credit-application/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/financial-engineering-expert/auto-experiment/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/financial-engineering-expert/auto-experiment/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/financial-engineering-expert/auto-experiment/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/financial-engineering-expert/auto-experiment/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/financial-engineering-expert/data-profiling/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/financial-engineering-expert/dnn-modeling/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/financial-engineering-expert/dnn-modeling/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/financial-engineering-expert/dnn-modeling/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/financial-engineering-expert/dnn-modeling/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/financial-engineering-expert/dnn-modeling/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedDianJin-SKILLS/financial-engineering-expert/dnn-tuning/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed
- No broad tool permissions
- No dangerous shell commands
- 149 network referencesin 41 files: social-media/scripts/xiaohongshu_publish.py:32 claim-adjustment/.env.example:4 client-kyc-profiling/SKILL.md:16 and 38 more
- No hidden or obfuscated text
- No instruction overrides
- 380 executable scriptsin 379 files: DianJin-SKILLS/financial-engineering-expert/xgb-modeling/scripts/modeling.py:1 claim-adjudication-review/scripts/validate_input.py:1 claim-adjustment/scripts/validate_input.py:1 and 376 more
- 158 references to secretsin 39 files: coverage-analysis/lib/llm_client.py:28 medical-review/lib/llm_client.py:28 claim-document-processing/scripts/check_completeness.py:12 and 36 more
- No requests to vote in Hot
Show all 687 signals by file
DianJin-SKILLS/claim-expert/claim-adjudication-review/.env.example
DianJin-SKILLS/claim-expert/claim-adjudication-review/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/claim-expert/claim-adjustment/.env.example
- Access to secretsline 2
[Credential-related source omitted] - Network accessline 4
# MCP服务器连接配置 CALCULATION_ENGINE_MCP_ENDPOINT=http://localhost:8080/mcp/calculation-engine - Network accessline 5
CALCULATION_ENGINE_MCP_ENDPOINT=http://localhost:8080/mcp/calculation-engine - Network accessline 6
CLAIMS_SYSTEM_MCP_ENDPOINT=http://localhost:8080/mcp/claims-system - Network accessline 7
POLICY_SYSTEM_MCP_ENDPOINT=http://localhost:8080/mcp/policy-system
DianJin-SKILLS/claim-expert/claim-adjustment/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/claim-expert/claim-case-registration/.env.example
DianJin-SKILLS/claim-expert/claim-case-registration/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/claim-expert/claim-document-processing/.env.example
DianJin-SKILLS/claim-expert/claim-document-processing/SKILL.md
DianJin-SKILLS/claim-expert/claim-document-processing/assets/analysis_output_schema.json
DianJin-SKILLS/claim-expert/claim-document-processing/references/reference.md
- Network accessline 192
[Credential-related source omitted] - Network accessline 193
[Credential-related source omitted] - Access to secretsline 193
[Credential-related source omitted] - Network accessline 194
base_url="https://dashscope.aliyuncs.com/compatible-mode/v1" - Access to secretsline 216
[Credential-related source omitted]
DianJin-SKILLS/claim-expert/claim-document-processing/scripts/analyze_materials.py
- Executable scriptline 1
#!/usr/bin/env python3 - Access to secretsline 17
[Credential-related source omitted] - Access to secretsline 179
[Credential-related source omitted] - Access to secretsline 189
[Credential-related source omitted] - Access to secretsline 222
[Credential-related source omitted] - Access to secretsline 405
[Credential-related source omitted]
DianJin-SKILLS/claim-expert/claim-document-processing/scripts/check_completeness.py
- Executable scriptline 1
#!/usr/bin/env python3 - Access to secretsline 12
[Credential-related source omitted] - Access to secretsline 179
[Credential-related source omitted] - Access to secretsline 190
[Credential-related source omitted] - Access to secretsline 220
[Credential-related source omitted] - Access to secretsline 249
[Credential-related source omitted] - Access to secretsline 354
[Credential-related source omitted] - Access to secretsline 365
[Credential-related source omitted]
DianJin-SKILLS/claim-expert/claim-document-processing/scripts/check_consistency.py
- Executable scriptline 1
#!/usr/bin/env python3 - Access to secretsline 13
[Credential-related source omitted] - Access to secretsline 242
[Credential-related source omitted] - Access to secretsline 252
[Credential-related source omitted] - Access to secretsline 283
[Credential-related source omitted] - Access to secretsline 310
[Credential-related source omitted] - Access to secretsline 428
[Credential-related source omitted] - Access to secretsline 439
[Credential-related source omitted]
DianJin-SKILLS/claim-expert/claim-document-processing/scripts/classify_claims.py
- Executable scriptline 1
#!/usr/bin/env python3 - Access to secretsline 10
[Credential-related source omitted] - Access to secretsline 11
[Credential-related source omitted] - Access to secretsline 605
[Credential-related source omitted] - Network accessline 612
[Credential-related source omitted] - Network accessline 613
[Credential-related source omitted] - Access to secretsline 613
[Credential-related source omitted] - Network accessline 614
base_url="https://dashscope.aliyuncs.com/compatible-mode/v1"
DianJin-SKILLS/claim-expert/claim-document-processing/scripts/cross_validate.py
- Executable scriptline 1
#!/usr/bin/env python3 - Access to secretsline 16
[Credential-related source omitted] - Access to secretsline 270
[Credential-related source omitted] - Access to secretsline 281
[Credential-related source omitted] - Access to secretsline 340
[Credential-related source omitted] - Access to secretsline 374
[Credential-related source omitted] - Access to secretsline 383
[Credential-related source omitted] - Access to secretsline 505
[Credential-related source omitted]
DianJin-SKILLS/claim-expert/claim-document-processing/scripts/utils.py
- Executable scriptline 1
#!/usr/bin/env python3 - Access to secretsline 197
[Credential-related source omitted] - Access to secretsline 208
[Credential-related source omitted] - Network accessline 221
[Credential-related source omitted] - Network accessline 222
[Credential-related source omitted] - Access to secretsline 222
[Credential-related source omitted] - Network accessline 223
base_url="https://dashscope.aliyuncs.com/compatible-mode/v1" - Access to secretsline 265
[Credential-related source omitted] - Access to secretsline 277
[Credential-related source omitted] - Access to secretsline 306
[Credential-related source omitted]
DianJin-SKILLS/claim-expert/claim-document-processing/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/claim-expert/claim-fraud-detection/.env.example
DianJin-SKILLS/claim-expert/claim-fraud-detection/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/claim-expert/claim-notification/.env.example
- Access to secretsline 2
[Credential-related source omitted] - Network accessline 4
# MCP服务器连接配置 NOTIFICATION_SYSTEM_MCP_ENDPOINT=http://localhost:8080/mcp/notification-system - Network accessline 5
NOTIFICATION_SYSTEM_MCP_ENDPOINT=http://localhost:8080/mcp/notification-system - Network accessline 6
CLAIMS_SYSTEM_MCP_ENDPOINT=http://localhost:8080/mcp/claims-system
DianJin-SKILLS/claim-expert/claim-notification/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/claim-expert/coverage-analysis/.env.example
DianJin-SKILLS/claim-expert/coverage-analysis/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/claim-expert/coverage-analysis/lib/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/claim-expert/coverage-analysis/lib/image_utils.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/claim-expert/coverage-analysis/lib/knowledge.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/claim-expert/coverage-analysis/lib/liability_exclusion.py
- Executable scriptline 1
Executable file - Access to secretsline 63
[Credential-related source omitted] - Access to secretsline 93
[Credential-related source omitted] - Access to secretsline 132
[Credential-related source omitted] - Access to secretsline 144
[Credential-related source omitted] - Access to secretsline 171
[Credential-related source omitted] - Access to secretsline 211
[Credential-related source omitted] - Access to secretsline 232
[Credential-related source omitted]
DianJin-SKILLS/claim-expert/coverage-analysis/lib/llm_client.py
- Executable scriptline 1
Executable file - Access to secretsline 28
[Credential-related source omitted] - Access to secretsline 31
[Credential-related source omitted] - Access to secretsline 32
[Credential-related source omitted] - Access to secretsline 37
[Credential-related source omitted] - Access to secretsline 42
[Credential-related source omitted] - Access to secretsline 43
[Credential-related source omitted] - Access to secretsline 46
[Credential-related source omitted] - Access to secretsline 47
[Credential-related source omitted] - Access to secretsline 48
[Credential-related source omitted] - Access to secretsline 51
[Credential-related source omitted] - Access to secretsline 59
[Credential-related source omitted] - Access to secretsline 60
[Credential-related source omitted] - Access to secretsline 62
[Credential-related source omitted] - Access to secretsline 71
[Credential-related source omitted] - Access to secretsline 72
[Credential-related source omitted] - Access to secretsline 79
[Credential-related source omitted] - Access to secretsline 87
[Credential-related source omitted] - Access to secretsline 102
[Credential-related source omitted] - Access to secretsline 118
[Credential-related source omitted] - Access to secretsline 134
[Credential-related source omitted] - Access to secretsline 144
[Credential-related source omitted] - Access to secretsline 161
[Credential-related source omitted] - Access to secretsline 174
[Credential-related source omitted] - Access to secretsline 200
[Credential-related source omitted] - Access to secretsline 207
[Credential-related source omitted] - Access to secretsline 222
[Credential-related source omitted] - Access to secretsline 228
[Credential-related source omitted]
DianJin-SKILLS/claim-expert/coverage-analysis/lib/prompts.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/claim-expert/coverage-analysis/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/claim-expert/medical-review/.env.example
DianJin-SKILLS/claim-expert/medical-review/lib/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/claim-expert/medical-review/lib/drug_indication.py
- Executable scriptline 1
Executable file - Access to secretsline 46
[Credential-related source omitted] - Access to secretsline 72
[Credential-related source omitted] - Access to secretsline 109
[Credential-related source omitted] - Access to secretsline 120
[Credential-related source omitted] - Access to secretsline 147
[Credential-related source omitted] - Access to secretsline 187
[Credential-related source omitted] - Access to secretsline 207
[Credential-related source omitted]
DianJin-SKILLS/claim-expert/medical-review/lib/image_utils.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/claim-expert/medical-review/lib/knowledge.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/claim-expert/medical-review/lib/llm_client.py
- Executable scriptline 1
Executable file - Access to secretsline 28
[Credential-related source omitted] - Access to secretsline 31
[Credential-related source omitted] - Access to secretsline 32
[Credential-related source omitted] - Access to secretsline 37
[Credential-related source omitted] - Access to secretsline 42
[Credential-related source omitted] - Access to secretsline 43
[Credential-related source omitted] - Access to secretsline 46
[Credential-related source omitted] - Access to secretsline 47
[Credential-related source omitted] - Access to secretsline 48
[Credential-related source omitted] - Access to secretsline 51
[Credential-related source omitted] - Access to secretsline 59
[Credential-related source omitted] - Access to secretsline 60
[Credential-related source omitted] - Access to secretsline 62
[Credential-related source omitted] - Access to secretsline 71
[Credential-related source omitted] - Access to secretsline 72
[Credential-related source omitted] - Access to secretsline 79
[Credential-related source omitted] - Access to secretsline 87
[Credential-related source omitted] - Access to secretsline 102
[Credential-related source omitted] - Access to secretsline 118
[Credential-related source omitted] - Access to secretsline 134
[Credential-related source omitted] - Access to secretsline 144
[Credential-related source omitted] - Access to secretsline 161
[Credential-related source omitted] - Access to secretsline 174
[Credential-related source omitted] - Access to secretsline 200
[Credential-related source omitted] - Access to secretsline 207
[Credential-related source omitted] - Access to secretsline 222
[Credential-related source omitted] - Access to secretsline 228
[Credential-related source omitted]
DianJin-SKILLS/claim-expert/medical-review/lib/prompts.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/claim-expert/medical-review/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/corporate-banker/credit-due-diligence/scripts/validate_due_diligence.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/corporate-banker/credit-industry-analysis/scripts/validate_industry.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/corporate-banker/equity-penetration-analysis/scripts/validate_equity.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/corporate-banker/financial-report-analysis/scripts/validate_financial_report.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/corporate-banker/pre-visit-plan/scripts/validate_visit_plan.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/corporate-banker/product-roadshow/scripts/validate_roadshow.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/corporate-banker/submit-credit-application/scripts/validate_credit_application.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/corporate-banker/visit-memo/scripts/validate_visit_memo.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-review-expert/admission-rules-scan/scripts/validate_admission_scan.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-review-expert/ai_risk_planning/scripts/validate_risk_plan.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-review-expert/credit-case-intake-check/scripts/validate_intake_check.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-review-expert/credit-collateral-risk-mgmt/references/registration-requirements.md
DianJin-SKILLS/credit-review-expert/credit-collateral-risk-mgmt/scripts/validate_collateral.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-review-expert/credit-related-party-detection/scripts/validate_related_party.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-review-expert/pre-visit-credit-analysis/scripts/validate_pre_visit.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-review-expert/reviewer-visit-memo/scripts/validate_visit_memo.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-risk-manager/credit-industry-rule-gen/scripts/validate_rule_output.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-risk-manager/credit-large-exposure-mgmt/scripts/validate_exposure.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-risk-manager/credit-policy-analysis/scripts/validate_policy_analysis.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-risk-manager/credit-risk-cot/scripts/validate_financial_data.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-risk-manager/credit-risk-extraction/scripts/validate_risk_extraction.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/credit-risk-manager/loan-risk-monitor/scripts/validate_monitor_report.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-risk-manager/post-loan-management/scripts/calculate_financial_ratios.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-risk-manager/post-loan-management/scripts/check_fund_usage.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-risk-manager/post-loan-management/scripts/evaluate_risk_classification.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-risk-manager/post-loan-management/scripts/validate_post_loan_data.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/credit-risk-manager/vlm-verifier/scripts/validate_fraud_report.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/data/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/data/loading.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/data/missing.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/data/splitting.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/data/woe_transform.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/data_utils.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/evaluation/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/evaluation/bootstrap.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/evaluation/metrics.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/evaluation/significance.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/report_artifact.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/result_emitter.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/xgb_core/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/xgb_core/evaluator.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/xgb_core/stability.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/_vendor/xgb_core/trainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/scripts/experiment_loop.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/scripts/hypothesis_generator.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/scripts/run_experiment.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/financial-engineering-expert/auto-experiment/scripts/session_tracker.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/auto-experiment/scripts/significance.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/data-profiling/_vendor/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/data-profiling/_vendor/result_emitter.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/data-profiling/scripts/profiler.py
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/_vendor/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/_vendor/data/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/_vendor/data/loading.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/_vendor/data/missing.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/_vendor/data/splitting.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/_vendor/data/woe_transform.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/_vendor/data_utils.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/_vendor/report_artifact.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/_vendor/result_emitter.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/_vendor/safety_gate.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/_vendor/xgb_core/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/_vendor/xgb_core/evaluator.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/_vendor/xgb_core/stability.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/_vendor/xgb_core/trainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-modeling/scripts/modeling.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/data/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/data/loading.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/data/missing.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/data/splitting.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/data/woe_transform.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/data_utils.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/report_artifact.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/result_emitter.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/tuning/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/tuning/diagnose.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/tuning/dnn_engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/tuning/engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/tuning/lr_engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/tuning/search_space.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/tuning/xgb_engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/xgb_core/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/xgb_core/evaluator.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/xgb_core/stability.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/_vendor/xgb_core/trainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/dnn-tuning/scripts/tuner.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/feature-analysis/_vendor/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/feature-analysis/_vendor/data/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/feature-analysis/_vendor/data/loading.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/feature-analysis/_vendor/data/missing.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/feature-analysis/_vendor/data/splitting.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/feature-analysis/_vendor/data/woe_transform.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/feature-analysis/_vendor/data_utils.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/feature-analysis/_vendor/fa_core/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/feature-analysis/_vendor/fa_core/correlation.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/feature-analysis/_vendor/fa_core/iv.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/feature-analysis/_vendor/fa_core/psi.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/feature-analysis/_vendor/fa_core/statistics.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/feature-analysis/_vendor/result_emitter.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/feature-analysis/scripts/analyzer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/data/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/data/loading.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/data/missing.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/data/splitting.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/data/woe_transform.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/data_utils.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/lr_cli.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/report_artifact.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/result_emitter.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/safety_gate.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/xgb_core/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/xgb_core/evaluator.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/xgb_core/stability.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/_vendor/xgb_core/trainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-modeling/scripts/modeling.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/data/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/data/loading.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/data/missing.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/data/splitting.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/data/woe_transform.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/data_utils.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/report_artifact.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/result_emitter.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/tuning/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/tuning/diagnose.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/tuning/dnn_engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/tuning/engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/tuning/lr_engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/tuning/search_space.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/tuning/xgb_engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/xgb_core/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/xgb_core/evaluator.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/xgb_core/stability.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/_vendor/xgb_core/trainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/lr-tuning/scripts/tuner.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/data/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/data/loading.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/data/missing.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/data/splitting.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/data/woe_transform.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/data_utils.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/evaluation/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/evaluation/bootstrap.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/evaluation/metrics.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/evaluation/significance.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/report_artifact.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/result_emitter.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/safety_gate.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/xgb_core/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/xgb_core/evaluator.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/xgb_core/stability.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/_vendor/xgb_core/trainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/analysis/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/analysis/agreement.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/analysis/feature_overlap.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/analysis/metrics.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/analysis/significance.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/analysis/stability.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/comparison.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/recommend/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/recommend/ranker.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/recommend/scenarios.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/reporting/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/reporting/builder.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/reporting/renderer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/reporting/sections.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/trainers/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/trainers/base.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/trainers/dnn_trainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/trainers/lr_trainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/trainers/xgb_trainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-comparison/scripts/tuned_params.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-explanation/_vendor/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-explanation/_vendor/result_emitter.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/model-explanation/scripts/explainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/_vendor/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/_vendor/data/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/_vendor/data/loading.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/_vendor/data/missing.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/_vendor/data/splitting.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/_vendor/data/woe_transform.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/_vendor/data_utils.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/_vendor/result_emitter.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/_vendor/xgb_core/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/_vendor/xgb_core/evaluator.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/_vendor/xgb_core/stability.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/_vendor/xgb_core/trainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/scripts/model_merger.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/scripts/run_segment.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/financial-engineering-expert/segment-modeling/scripts/segment_explorer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/scripts/segment_hypothesis.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/segment-modeling/scripts/segment_strategies.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/univariate-analysis/_vendor/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/univariate-analysis/_vendor/data/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/univariate-analysis/_vendor/data/loading.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/univariate-analysis/_vendor/data/missing.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/univariate-analysis/_vendor/data/splitting.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/univariate-analysis/_vendor/data/woe_transform.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/univariate-analysis/_vendor/data_utils.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/univariate-analysis/_vendor/result_emitter.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/univariate-analysis/scripts/analyzer.py
- Executable scriptline 1
#!/usr/bin/env python
DianJin-SKILLS/financial-engineering-expert/univariate-analysis/scripts/binning.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/univariate-analysis/scripts/cross_distribution.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/data/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/data/loading.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/data/missing.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/data/splitting.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/data/woe_transform.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/data_utils.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/report_artifact.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/result_emitter.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/safety_gate.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/tuning/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/tuning/diagnose.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/tuning/dnn_engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/tuning/engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/tuning/lr_engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/tuning/search_space.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/tuning/xgb_engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/tuning_engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/xgb_cli.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/xgb_core/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/xgb_core/evaluator.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/xgb_core/stability.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/_vendor/xgb_core/trainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/scripts/comparator.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/scripts/stacker.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-deepmodel/scripts/sub_trainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/data/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/data/loading.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/data/missing.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/data/splitting.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/data/woe_transform.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/data_utils.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/diagnostic/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/diagnostic/fit_check.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/diagnostic/ks_structure.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/diagnostic/report.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/fa_core/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/fa_core/correlation.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/fa_core/iv.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/fa_core/psi.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/fa_core/statistics.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/model_card.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/report_artifact.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/result_emitter.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/safety_gate.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/tuning/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/tuning/diagnose.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/tuning/search_space.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/xgb_cli.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/xgb_core/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/xgb_core/evaluator.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/xgb_core/stability.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/_vendor/xgb_core/trainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/scripts/feature_selector.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-modeling/scripts/modeling.py
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/data/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/data/loading.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/data/missing.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/data/splitting.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/data/woe_transform.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/data_utils.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/diagnostic/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/diagnostic/fit_check.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/diagnostic/ks_structure.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/diagnostic/report.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/model_card.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/report_artifact.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/result_emitter.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/safety_gate.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/tuning/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/tuning/diagnose.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/tuning/dnn_engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/tuning/engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/tuning/lr_engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/tuning/search_space.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/tuning/xgb_engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/tuning_engine.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/xgb_cli.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/xgb_core/__init__.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/xgb_core/evaluator.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/xgb_core/stability.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/_vendor/xgb_core/trainer.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/financial-engineering-expert/xgb-tuning/scripts/tuner.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/insurance-agent/activity-planner/.env.example
DianJin-SKILLS/insurance-agent/activity-planner/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/application-check/.env.example
DianJin-SKILLS/insurance-agent/application-check/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/client-consulting/.env.example
DianJin-SKILLS/insurance-agent/client-consulting/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/customer-profiling/.env.example
DianJin-SKILLS/insurance-agent/customer-profiling/scripts/analyze_profile.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/customer-profiling/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/gap-advisor/.env.example
DianJin-SKILLS/insurance-agent/gap-advisor/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/health-disclosure-guide/.env.example
DianJin-SKILLS/insurance-agent/health-disclosure-guide/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/objection-handler/.env.example
DianJin-SKILLS/insurance-agent/objection-handler/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/plan-generator/.env.example
DianJin-SKILLS/insurance-agent/plan-generator/lib/benefit_calculator.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/insurance-agent/plan-generator/lib/generate_plan.py
- Executable scriptline 1
Executable file
DianJin-SKILLS/insurance-agent/plan-generator/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/policy-explanation/.env.example
DianJin-SKILLS/insurance-agent/policy-explanation/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/product-wiki/.env.example
DianJin-SKILLS/insurance-agent/product-wiki/install.sh
- Executable scriptline 1
#!/usr/bin/env bash
DianJin-SKILLS/insurance-agent/product-wiki/scripts/extract_doc.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/product-wiki/scripts/test_skill_v2.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/product-wiki/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/product-wiki/uninstall.sh
- Executable scriptline 1
#!/usr/bin/env bash
DianJin-SKILLS/insurance-agent/renewal-assist/.env.example
DianJin-SKILLS/insurance-agent/renewal-assist/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/social-media/.env.example
DianJin-SKILLS/insurance-agent/social-media/SKILL.md
- Access to secretsline 59
[Credential-related source omitted]
DianJin-SKILLS/insurance-agent/social-media/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/insurance-agent/social-media/scripts/xiaohongshu_publish.py
- Executable scriptline 1
Executable file - Network accessline 32
CREATOR_URL = "https://creator.xiaohongshu.com/publish/publish" - Network accessline 58
await page.goto("https://creator.xiaohongshu.com/login") - Network accessline 72
if "www.xiaohongshu.com" in current_url and "login" not in current_url: print("[检测] 登录成功，正在跳转创作者中心...") await page.goto("https://creator.xiaohongshu.com/publish/publish") - Network accessline 73
print("[检测] 登录成功，正在跳转创作者中心...") await page.goto("https://creator.xiaohongshu.com/publish/publish") await asyncio.sleep(3) - Network accessline 74
await page.goto("https://creator.xiaohongshu.com/publish/publish")
DianJin-SKILLS/investment-researcher/alternative-factor-mining/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/announcement-analysis/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/company-deep-analysis/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/company-deep-analysis/output/五粮液深度分析报告.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/company-deep-analysis/output/贵州茅台深度分析报告.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/company-one-page-analysis/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/convertible-bond-valuation/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/earnings-commentary-generator/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/fixed-income-daily-brief/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/fixed-income-overnight-brief/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/fixed-income-rate-review/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/fixed-income-weekly-brief/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/global-macro-linkage/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/industry-bond-risk-control/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/industry-deep-analysis/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/industry-one-page-analysis/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/institutional-research-outline/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/lgt-debt-risk-analysis/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/macro-asset-allocation/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/macro-daily-briefing/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/macro-data-brief/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/macro-policy-analysis/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/macro-risk-monitor/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/macro-weekly-briefing/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/market-minimal-review/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/market-sentiment-monitor/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/market-trend-style-analysis/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/policy-flash-briefing/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/quant-daily-brief/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/quant-factor-tracker/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/quant-strategy-quick-backtest/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/quant-weekly-brief/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/sector-allocation/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/strategy-daily-briefing/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/investment-researcher/valuation-prosperity-tracking/SKILL.md
- Executable scriptline 1
Executable file
DianJin-SKILLS/underwriting-expert/conclusion-interpretation/.env.example
DianJin-SKILLS/underwriting-expert/conclusion-interpretation/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/underwriting-expert/follow-up-outreach/.env.example
DianJin-SKILLS/underwriting-expert/follow-up-outreach/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/underwriting-expert/health-verification/.env.example
DianJin-SKILLS/underwriting-expert/health-verification/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/underwriting-expert/medical-assessment/.env.example
DianJin-SKILLS/underwriting-expert/medical-assessment/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/underwriting-expert/recording-inspection/.env.example
DianJin-SKILLS/underwriting-expert/recording-inspection/scripts/validate_input.py
- Executable scriptline 1
#!/usr/bin/env python3
DianJin-SKILLS/wealth-copilot/L2-1_opportunity/client-kyc-profiling/SKILL.md
- Network accessline 16
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 17
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 21
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 22
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-1_opportunity/client-opportunity-scan/SKILL.md
- Network accessline 16
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 17
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 21
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 22
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-1_opportunity/maturity-client-alert/SKILL.md
- Network accessline 16
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 17
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 21
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 22
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-2_outreach/client-care-script/SKILL.md
- Network accessline 16
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 17
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 21
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 22
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-2_outreach/hot-product-briefing/SKILL.md
- Network accessline 16
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 17
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 21
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 22
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-2_outreach/market-event-interpreter/SKILL.md
- Network accessline 18
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 19
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 23
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 24
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-2_outreach/market-morning-brief/SKILL.md
- Network accessline 16
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 17
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 21
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 22
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-2_outreach/outreach-script-generator/SKILL.md
- Network accessline 16
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 17
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 21
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 22
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-2_outreach/wechat-moments-creator/SKILL.md
- Network accessline 16
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 17
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 21
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 22
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-3_strategy/client-segmentation-strategy/SKILL.md
- Network accessline 16
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 17
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 21
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 22
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-3_strategy/competitor-product-compare/SKILL.md
- Network accessline 17
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 18
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 22
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 23
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-3_strategy/sales-sop-navigator/SKILL.md
- Network accessline 17
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 18
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 22
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 23
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-4_qa/compliance-risk-qa/SKILL.md
- Network accessline 17
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 18
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 22
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 23
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-4_qa/market-insight-qa/SKILL.md
- Network accessline 17
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 18
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 22
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 23
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-4_qa/product-knowledge-qa/SKILL.md
- Network accessline 17
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 18
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 22
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 23
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-5_diagnosis/fund-deep-research/SKILL.md
- Network accessline 16
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 17
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 21
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 22
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-5_diagnosis/portfolio-health-check/SKILL.md
- Network accessline 17
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 18
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 22
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 23
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-5_diagnosis/portfolio-risk-radar/SKILL.md
- Network accessline 17
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 18
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 22
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 23
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-6_allocation/asset-allocation-optimizer/SKILL.md
- Network accessline 17
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 18
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 22
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 23
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-6_allocation/family-financial-planner/SKILL.md
- Network accessline 12
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 13
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 17
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 18
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-6_allocation/investment-simulation/SKILL.md
- Network accessline 17
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 18
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 22
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 23
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-6_allocation/smart-product-matching/SKILL.md
- Network accessline 17
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 18
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 22
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 23
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-7_summary/daily-work-review/SKILL.md
- Network accessline 16
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 17
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 21
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 22
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-7_summary/service-summary-generator/SKILL.md
- Network accessline 16
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 17
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 21
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 22
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-7_summary/tomorrow-plan-generator/SKILL.md
- Network accessline 17
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 18
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 22
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 23
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-8_companion/investor-education-qa/SKILL.md
- Network accessline 16
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 17
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 21
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 22
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-8_companion/market-hotspot-digest/SKILL.md
- Network accessline 18
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 19
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 23
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 24
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/L2-8_companion/portfolio-alert-narrator/SKILL.md
- Network accessline 17
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 18
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 22
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 23
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
DianJin-SKILLS/wealth-copilot/_template/SKILL.md
- Network accessline 14
**盈米金融数据（qieman）** - 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 15
- 服务地址：`https://dashscope.aliyuncs.com/api/v1/mcps/Qieman/sse` - Network accessline 19
**恒生聚源金融数据（gildata-aidata）** - 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp` - Network accessline 20
- 服务地址：开通恒生聚源 MCP 服务后获取，格式为 `https://dashscope.aliyuncs.com/api/v1/mcps/<your-mcp-id>/mcp`
How the signals fit together
Facts from a static read of the files at the pinned commit. They describe what the code does, not why.
Some files could not be read; the facts below still hold.
Declared and executed: 1 fact
- Contacts creator.xiaohongshu.com, which SKILL.md, the README and the plugin manifest do not mentionsocial-media/scripts/xiaohongshu_publish.py:32High confidence
Signal confidence: 499 facts
- Executable script in code of a file that runsclaim-adjudication-review/scripts/validate_input.py:1High confidence
- Executable script in code of a file that runsclaim-adjustment/scripts/validate_input.py:1High confidence
- Executable script in code of a file that runsclaim-case-registration/scripts/validate_input.py:1High confidence
- Access to secrets in code of the SKILL.mdclaim-document-processing/SKILL.md:47High confidence
- Access to secrets in code of the SKILL.md: a variable, placeholder or call, not a stored valueclaim-document-processing/SKILL.md:49Low confidence
- Network access in code of documentationclaim-document-processing/references/reference.md:192Low confidence
- Network access in code of documentationclaim-document-processing/references/reference.md:193Low confidence
- Access to secrets in code of documentation: a variable, placeholder or call, not a stored valueclaim-document-processing/references/reference.md:193Low confidence
- Network access in code of documentationclaim-document-processing/references/reference.md:194Low confidence
- Access to secrets in code of documentation: a variable, placeholder or call, not a stored valueclaim-document-processing/references/reference.md:216Low confidence
- Executable script in code of a file that runsclaim-document-processing/scripts/analyze_materials.py:1High confidence
- Access to secrets in prose of a file that runs: a variable, placeholder or call, not a stored valueclaim-document-processing/scripts/analyze_materials.py:17Low confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/analyze_materials.py:179High confidence
- Access to secrets in prose of a file that runsclaim-document-processing/scripts/analyze_materials.py:189Low confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/analyze_materials.py:222High confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/analyze_materials.py:405High confidence
- Executable script in code of a file that runsclaim-document-processing/scripts/check_completeness.py:1High confidence
- Access to secrets in prose of a file that runs: a variable, placeholder or call, not a stored valueclaim-document-processing/scripts/check_completeness.py:12Low confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/check_completeness.py:179High confidence
- Access to secrets in prose of a file that runsclaim-document-processing/scripts/check_completeness.py:190Low confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/check_completeness.py:220High confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/check_completeness.py:249High confidence
- Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored valueclaim-document-processing/scripts/check_completeness.py:354Low confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/check_completeness.py:365High confidence
- Executable script in code of a file that runsclaim-document-processing/scripts/check_consistency.py:1High confidence
- Access to secrets in prose of a file that runs: a variable, placeholder or call, not a stored valueclaim-document-processing/scripts/check_consistency.py:13Low confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/check_consistency.py:242High confidence
- Access to secrets in prose of a file that runsclaim-document-processing/scripts/check_consistency.py:252Low confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/check_consistency.py:283High confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/check_consistency.py:310High confidence
- Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored valueclaim-document-processing/scripts/check_consistency.py:428Low confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/check_consistency.py:439High confidence
- Executable script in code of a file that runsclaim-document-processing/scripts/classify_claims.py:1High confidence
- Access to secrets in prose of a file that runs: a variable, placeholder or call, not a stored valueclaim-document-processing/scripts/classify_claims.py:10Low confidence
- Access to secrets in prose of a file that runs: a variable, placeholder or call, not a stored valueclaim-document-processing/scripts/classify_claims.py:11Low confidence
- Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored valueclaim-document-processing/scripts/classify_claims.py:605Low confidence
- Network access in code of a file that runsclaim-document-processing/scripts/classify_claims.py:612High confidence
- Network access in code of a file that runsclaim-document-processing/scripts/classify_claims.py:613High confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/classify_claims.py:613High confidence
- Network access in code of a file that runsclaim-document-processing/scripts/classify_claims.py:614High confidence
- Executable script in code of a file that runsclaim-document-processing/scripts/cross_validate.py:1High confidence
- Access to secrets in prose of a file that runs: a variable, placeholder or call, not a stored valueclaim-document-processing/scripts/cross_validate.py:16Low confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/cross_validate.py:270High confidence
- Access to secrets in prose of a file that runsclaim-document-processing/scripts/cross_validate.py:281Low confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/cross_validate.py:340High confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/cross_validate.py:374High confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/cross_validate.py:383High confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/cross_validate.py:505High confidence
- Executable script in code of a file that runsclaim-document-processing/scripts/utils.py:1High confidence
- Access to secrets in code of a file that runsclaim-document-processing/scripts/utils.py:197High confidence
and 449 more
Only the first 500 facts are listed.
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Compatible | Standard SKILL.md DianJin-SKILLS/claim-expert/claim-adjudication-review/SKILL.md | .claude/skills/<skill-name>/ |
| Codex | Compatible | Standard SKILL.md DianJin-SKILLS/claim-expert/claim-adjudication-review/SKILL.md | .agents/skills/<skill-name>/ |
| Cursor | Compatible | Standard SKILL.md DianJin-SKILLS/claim-expert/claim-adjudication-review/SKILL.md | .cursor/skills/<skill-name>/ |
| Gemini CLI | Compatible | Standard SKILL.md DianJin-SKILLS/claim-expert/claim-adjudication-review/SKILL.md | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Compatible | Standard SKILL.md DianJin-SKILLS/claim-expert/claim-adjudication-review/SKILL.md | .github/skills/<skill-name>/ |
| OpenCode | Compatible | Standard SKILL.md DianJin-SKILLS/claim-expert/claim-adjudication-review/SKILL.md | .opencode/skills/<skill-name>/ |
- Native The repository ships a manifest, folder or frontmatter field for this agent.
- Compatible A standard SKILL.md that this agent can read.
- Unknown No evidence either way.
Derived automatically from paths, manifests and frontmatter. Not tested on each agent.
