菜单
Canonical conventions for the forensic accounting ecosystem — preloaded into convention-auditor agent as the single source of truth for convention checks.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Scan all ecosystem repos for convention deviations. Reports findings; does not fix anything. Use proactively after cross-repo changes.
Validates pipeline parquet outputs for schema compliance, null rates, row count baselines, and freshness. Use after pipeline runs or when triage shows data quality concerns. Fast and read-only (uses haiku).
Run tests across all ecosystem repos and report pass/fail summary. Use when the user wants to verify the ecosystem is healthy or after cross-repo changes.
Runs the krff-shell ETL pipeline with pre/post validation. Use when data needs refreshing, after dependency updates, or when triage shows stale parquets. Reads CLAUDE.md before running.
Research across sibling repos to gather context for cross-project tasks. Use when working in one repo and needing to understand how another repo works, what it exports, or how data flows between them.
Show all open work (board + backlog) with dependency-aware execution order
| name | canonical-conventions |
| description | Canonical conventions for the forensic accounting ecosystem — preloaded into convention-auditor agent as the single source of truth for convention checks. |
| user-invocable | false |
This is the authoritative convention table for the Korean forensic accounting ecosystem. The convention-auditor agent uses this to check all repos for deviations.
| # | Convention | Expected | Check Method | Exceptions |
|---|---|---|---|---|
| 1 | Build system | hatchling (requires = ["hatchling"] in [build-system]) | grep build-backend pyproject.toml | jfia-catalog (data artifact, no build system required) |
| 2 | Python version | >=3.11 in [project] requires-python | grep requires-python pyproject.toml | kr-beneish: >=3.10 is acceptable |
| 3 | Package manager | uv (never pip) | Check CLAUDE.md mentions uv; no requirements.txt present | — |
| 4 | Test invocation | uv run pytest tests/ -v | Check CLAUDE.md for test command | kr-company-registry: bare pytest tests/ -v is acceptable (no uv wrapper needed) |
| 5 | uv.lock committed | File tracked in git | git -C <repo> ls-files uv.lock | jfia-catalog (no dependencies) |
| 6 | conftest.py present | Required in repos that have a tests/ directory | test -f tests/conftest.py | Repos without tests/ are exempt |
| 7 | constants.py present | Required in repos with magic strings/thresholds in code | Check for src/*/constants.py or src/constants.py | kr-trading-calendar (no magic strings beyond holiday list) |
| 8 | Paths module | _paths.py or paths.py in src | find src/ -name '*paths*' | Repos with no file I/O are exempt |
| 9 | Commit message style | Conventional prefix: feat/fix/docs/refactor/test/chore | git log --oneline -5 | — |
| 10 | .claude/ directory | Present at repo root | test -d .claude | jfia-catalog (minimal repo) |
| 11 | compile-bytecode | compile-bytecode = false in [tool.uv] | grep compile-bytecode pyproject.toml | Repos without pyproject.toml |
| 12 | CLAUDE.md present | Present at repo root | test -f CLAUDE.md | — |
| 13 | Known Gaps section | ## Known Gaps table in CLAUDE.md | grep '## Known Gaps' CLAUDE.md | Hub exempt (coordinator, not code) |
| 14 | No stale repo-name references | Zero matches for kr-forensic-finance in .md, .toml, .conf files | grep -rl "kr-forensic-finance" --include="*.md" --include="*.toml" --include="*.conf" | Files under reports/ (historical session logs) |
All repos under C:\Users\pon00\Projects\:
krff-shell (delivery shell — renamed from kr-forensic-finance)kr-forensic-corekr-dart-pipelinekr-anomaly-scoringkr-stat-testskr-company-registrykr-trading-calendarkr-beneishkr-derivativesjfia-catalogjfia-forensickr-enforcement-caseskr-real-estate