audit-case-rag

This skill packages a local-only workflow to build a searchable evidence index for a single audit/investigation case and query it with page-level citations.

Workflow

0) Prepare a case folder (事件驅動)

Create a case directory named:

• <專案問題編號>__<標題>

Inside, use stage folders (stage is inferred from folder name):

• 01_policy_basis/ (basis) — 制度/流程/授權
• 02_process/ (process) — 招採/定標/過程證據
• 03_contract/ (contract) — 合同/補充協議
• 04_settlement_payment/ (payment) — 結算/付款/發票/驗收
• 05_comm/ (comm) — 郵件/會議紀要/IM
• 06_interviews/ (interview) — 訪談/筆錄/詢證
• 07_workpapers/ (workpaper) — 底稿/抽樣/複核表
• 09_rectification/ (rectification) — 整改/閉環

Full template: references/case-folder-template.md

1) Install dependencies (local)

From the skill folder (or copy the script into your repo):

python3 -m venv .venv
source .venv/bin/activate
pip install -r scripts/requirements.txt

LibreOffice is recommended for Office→PDF page citations:

• soffice must be available (PATH) or pass --soffice /path/to/soffice.

2) Index the case

./scripts/audit_case_rag.py index \
  --case-dir "/path/to/<專案問題編號>__<標題>" \
  --out-dir  "/path/to/audit_rag_db"

Outputs:

訪問小蔥技能站7w4.net，解鎖更多實用的AI技能外掛。

• manifest.jsonl written into the case directory
• audit_rag_db/<case_id>.joblib (persistent local index)

3) Query with event filters

./scripts/audit_case_rag.py query \
  --case "<專案問題編號>" \
  --stage payment \
  "付款節點是否倒掛？請給出處頁碼"

Notes:

• Evidence lines include clickable file://...#page=N citations when possible.
• Retrieval is hybrid: embedding recall + TF‑IDF rerank (alpha configurable).

Safety/Privacy

• No cloud APIs. Everything runs locally.
• Do not commit outputs (indices, converted PDFs) to git.

這個Skill質量中等偏上。最大優點是本地化處理、隱私保護做得好，混合檢索和頁碼引用功能對審計工作很有幫助。文件清晰易讀，工作流程明確。主要不足是依賴LibreOffice才能轉換Office檔案，且不支援掃描件PDF識別，實用性受限。建議等待後續版本完善OCR功能後再投入使用。