Individual skill
erp-finance-review
Solve finance-review tasks against the shared ERP/compliance API, including reimbursement/AP close batches, stale AP snapshot reconciliation, vendor onboarding release control, account-change payment release reviews, and prepaid close reconciliations. Use when a prompt provides a TASK_ENV_BASE_URL plus local payload templates/scope files and asks for a JSON decision or close report from current API records.
experiments/codex_skill_creator_comparison_gpt5_5_xhigh/reports/skills/task_group_005/fewshot/naive/fewshot_attempt_02/SKILL.md
Source commit 5082a777379156a048881e94940d640702e5e09a
What was checked
- Matches the observed repository HEAD
- Yes
- Scanner rules are current
- No
- Repository scan complete
- No
- Skill scan complete
- No
- Observed behavior
- Not checked yet
- Last checked
- Nov 6, 2026
These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.
Install this skill
Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.
An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.
Repository context
- Repository stars (not skill ratings)
- 121
- Repository last updated
- Oct 29, 2026
- License
- Apache-2.0
Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.
- agent Repository topic
- ai Repository topic
- benchmark Repository topic
- llm Repository topic
- self-evolving Repository topic
- self-learning Repository topic
- skills Repository topic
- legal Derived category
- finance Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
Read excerpt
# ERP Finance Review
## Core Workflow
1. Read the prompt, answer template, and all local payload files for the task. Extract the requested IDs, period/as-of date, entity, account scope, thresholds, and required output ordering.
2. Treat the API as the system of record. Use local CSV/JSON payloads only for scope, stale context, and output schema.
3. Use the runner-provided base URL. Prefer `/api/...` endpoints, falling back to non-`/api` variants named in the prompt. `/endpoints` lists available paths; exact-match query parameters by field name are supported.
4. Query only records needed for the scoped IDs. Paginate with `limit` and `offset` if a collection result reports `total > count`.
5. Preserve the answer template exactly: required keys only when `additional_properties_allowed` is false, correct enum values, requested top-level order, and list ordering.
6. Use decimal arithmetic for money. Report USD amounts to two decimals unless the template asks for cents. Compare money after rounding to cents, but keep source amounts as authoritative.
7. Return JSON only unless the user explicitly requests explanation.
## API Map
