Individual skill
harborcrm-reconciliation
Solve HarborCRM API-backed CRM handoff, trade-show prospecting, and import-batch cleanup tasks that require exact JSON output from an answer template.
experiments/codex_skill_creator_comparison_gpt5_5_xhigh/reports/skills/task_group_001/fewshot/naive/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.
- agent Repository topic
- ai Repository topic
- benchmark Repository topic
- llm Repository topic
- self-evolving Repository topic
- self-learning Repository topic
- skills Repository topic
- sales 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
Use this skill when a task asks for a HarborCRM JSON handoff or prospecting summary using a runner-provided API base URL and an `input/payloads/answer_template.json` file.
## Core Procedure
1. Read the prompt and answer template before fetching data. Treat the template as the output contract: required top-level keys, nested keys, enum values, null-vs-empty-string expectations, ordering rules, and integer/date precision all come from the template.
2. Determine the API base URL supplied by the runner. Query only public business endpoints described by the prompt or exposed by the task environment.
3. Fetch every relevant collection before deciding. For event tasks this usually means event details, sponsor orders or sponsor packages, badges, finance invoices, CRM accounts, CRM contacts, opportunities, campaign members, and policies. For trade-show tasks this usually means show details, exhibitors, meeting-interest records, CRM accounts, CRM contacts, and policies. For import-batch tasks this usually means batch metadata, raw contacts, suppression records, CRM accounts, CRM contacts, and policies.
4. Join records deterministically. Prefer stable IDs when present; otherwise match companies by normalized company name and contacts by normalized email, then normalized phone, then exact contact name within the matched account.
5. Build the JSON from facts already fetched. Do not add explanatory prose, template metadata, comments, or undeclared fields.
6. Validate the completed object against the template shape, enum values, counts, totals, sorting rules, and JSON syntax before returning it.
Useful shell pattern:
```bash