Individual skill
harborcrm-json-handoff
Solve HarborCRM API tasks that ask Codex to produce CRM-ready JSON handoffs from event, trade-show, sponsor, finance, badge, import-batch, CRM account, contact, campaign-member, policy, or meeting-interest data. Use when a prompt references HarborCRM, answer_template.json, CRM import cleanup, event sponsor or badge reconciliation, trade-show prospect qualification, duplicate or suppression handling, lead handoff, or campaign-member action summaries.
experiments/codex_skill_creator_comparison_gpt5_5_xhigh/reports/skills/task_group_001/fewshot/codex/fewshot_attempt_01/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.
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
SKILL.mdcore · 10,190 bytesSKILL.mdcore · 10,190 bytesSKILL.mdcore · 10,190 bytesSKILL.mdcore · 10,190 bytesSKILL.mdcore · 10,190 bytesagents/openai.yamlasset · 240 bytesscripts/harborcrm_helpers.pyruns · 4,281 bytes
Read excerpt
# HarborCRM JSON Handoff
## Core Workflow
1. Read the user prompt and `input/payloads/answer_template.json` first. Treat the template as the response contract: preserve required top-level keys, field names, enum spellings, `null` versus empty string conventions, integer currency/counts, and sorting rules. Return one JSON object only.
2. Query the runner-provided HarborCRM base URL. Use the endpoint list from the prompt or environment-access file; if a route is unavailable, continue with the available related routes and do not invent records.
3. Build normalized lookup tables before reasoning:
- Accounts by `account_id`, normalized company name, and domain.
- Contacts by `contact_id`, normalized email, digits-only phone, and account plus normalized contact name.
- Opportunities by event/show/account when those fields exist.
- Campaign members by event/campaign plus account/contact pair.
- Event sponsors, orders, packages, invoices, badges, exhibitors, meeting-interest rows, raw import rows, and suppression rows by their natural IDs plus normalized company/contact keys.
4. Prefer explicit policy records and prompt-specific rules over defaults. Use defaults below only when the task asks for the same HarborCRM pattern and no policy field overrides it.
5. Assemble the JSON directly from derived facts, then validate it against the template: no prose, no undeclared fields, sorted arrays, exact enum values, and complete aggregate totals.