Individual skill

erp-finance-api-reconciler

Use this skill for task-group ERP finance/compliance reconciliation prompts that require querying a runner-provided JSON API and returning exact JSON. It is especially relevant for reimbursement-to-AP close batches, stale AP snapshot reviews, prepaid close checks, vendor onboarding release decisions, account-change payment release reviews, and prompts mentioning claims, AP bills, payments, close logs, vendors, compliance objects, ownership, registry, screening, bank checks, prepaid invoices, or GL balances.

experiments/codex_skill_creator_comparison_gpt5_5_xhigh/reports/skills/task_group_005/fewshot/opencode/fewshot_attempt_02/SKILL.md

Read pinned source

Source commit 5082a777379156a048881e94940d640702e5e09a

What was checked

Matches the observed repository HEAD
Yes
Scanner rules are current
Yes
Repository scan complete
No
Skill scan complete
No
Observed behavior
Instructions only
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
Read excerpt

# ERP Finance API Reconciler

Use this skill when a task asks for finance/compliance release or close decisions from a shared ERP-style API. The work is evidence reconciliation, not summarization: local payloads define scope and output shape, while the current API records are the system of record.

## Core Workflow

1. Read the prompt and every local payload. Treat `answer_template.json` as the output contract: top-level keys, enum values, ordering, required constants, and numeric precision come from the template.
2. Get the API base URL from the runner placeholder or environment. Do not hardcode training URLs. Discover `/endpoints` when useful.
3. Pull current records for only the scoped IDs, accounts, periods, and entities. The API supports exact-match query parameters by field name plus `limit` and `offset`; paginate until all relevant records are available.
4. Build a short evidence table before finalizing: one row per scoped claim, business, invoice, or account, with the source fields that drive each decision.
5. Produce only the requested JSON. Do not include narrative text, citations, or extra keys unless the template asks for them.

When a prompt says "USD cents" in this task family, interpret it as USD amounts reported to cent precision unless the template explicitly requires integer cents.