pack-builder

分类: 工具与效率 | 上传者: yang478 | 下载: 0 | 版本: v1.0（最新）

Use when generating an auditable, deterministic knowledge pack from one or more documents (txt/md/docx/readable-pdf), producing `references/` + `kb.sqlite` (FTS5, no embeddings) + a `kbtool` CLI (root wrapper + python scripts + optional per-platform binary) for deterministic search→context bundling and citations.

更新日志: Source: GitHub https://github.com/yang478/auditable-knowledge-packs

目录结构

当前层级: pack-builder/

📁 scripts/
- 📁 build_skill_lib/
  - 📄 __init__.py 451 B
  - 📄 build.py 7.8 KB
  - 📄 catalog.py 4.8 KB
  - 📄 cli.py 3.2 KB
  - 📄 extract.py 3.7 KB
  - 📄 fs_utils.py 3.4 KB
  - 📄 indexes.py 2.9 KB
  - 📄 kbtool_assets.py 8.1 KB
  - 📄 references.py 31.9 KB
  - 📄 render.py 6.1 KB
  - 📄 sqlite_db.py 17.2 KB
  - 📄 text_utils.py 4.9 KB
  - 📄 types.py 1.7 KB
- 📁 tests/
  - 📁 fixtures/
    
    📁 retrieval/
    
    📄 handbook.md 328 B
    
    📄 standard_v1.md 657 B
    
    📄 standard_v2.md 663 B
  - 📄 test_build_skill.py 80.9 KB
  - 📄 test_llm_catalog.py 13.5 KB
- 📄 build_skill.py 280 B
- 📄 llm_catalog.py 22.9 KB
📁 templates/
- 📁 kbtool_lib/
  - 📄 __init__.py 36 B
  - 📄 catalog.py 6.8 KB
  - 📄 cli.py 7.7 KB
  - 📄 reindex.py 30.9 KB
  - 📄 retrieval.py 32.3 KB
  - 📄 runtime.py 2.3 KB
  - 📄 text.py 6.3 KB
- 📄 kbtool.py 174 B
- 📄 reindex.py 5.0 KB
- 📄 taxonomy.doctype.v1.json 1.7 KB
- 📄 taxonomy.domain.v1.json 2.6 KB
📄 SKILL.md 5.2 KB

SKILL.md

---
name: pack-builder
description: Use when generating an auditable, deterministic knowledge pack from one or more documents (txt/md/docx/readable-pdf), producing `references/` + `kb.sqlite` (FTS5, no embeddings) + a `kbtool` CLI (root wrapper + python scripts + optional per-platform binary) for deterministic search→context bundling and citations.
---

# Auditable Knowledge Pack Builder

Generate a `monitor`-style knowledge base skill from one or more documents:

- Progressive disclosure layout: `references/<doc_id>/{metadata.md,toc.md,chapters/,sections/}`
- SQLite index for fast non-vector search: `kb.sqlite` (CJK 2-gram + ASCII word tokens in FTS5)
- Deterministic bundle command (recommended): `./kbtool bundle` → outputs a single `bundle.md` with forced sources
- Optional sharded TSV indexes: `indexes/headings/*.tsv`, `indexes/kw/*.tsv` (fallback only)
  - `scripts/kbtool.py` and `scripts/kbtool_lib/*.py` are the python implementation
  - `bin/<platform>/kbtool(.exe)` is optional (PyInstaller); root `kbtool` wrapper prefers a fresh matching binary

## Controlled Retrieval V1

- 在线主流程固定为：标题/正文/术语 三路召回 → 确定性融合排序 → 最多一轮补查 → `bundle.md`
- 一轮补查只允许 `definition`、`references`、`version_metadata` 这类受控动作
- `./kbtool reindex`（或 `python3 scripts/kbtool.py reindex`）使用 shadow rebuild + 原子重建（atomic switch）来激活新版本，并保留旧版本文档行为非激活记录

## Quick Start

1. Choose an output skill name (lowercase letters/digits/hyphens only), e.g. `my-books`.
2. Run:
   - From this repo: `python3 pack-builder/scripts/build_skill.py --skill-name my-books --inputs /path/to/book1.pdf /path/to/book2.docx`
   - If installed under `.claude/skills`: `python3 .claude/skills/pack-builder/scripts/build_skill.py --skill-name my-books --inputs ...`
3. Use the generated skill at `.claude/skills/my-books/`.

## Command Reference

- Show help: `python3 .claude/skills/pack-builder/scripts/build_skill.py --help`
- Write to a specific directory: `--out-dir .claude/skills`
- Overwrite an existing output folder: `--force`

## What You Provide

- One file or many files via `--inputs` (supports `.md`, `.txt`, `.docx`, readable `.pdf`)
- Optional `--title` for the generated skill’s human-friendly heading

## Output Layout (Generated Skill)

```
.claude/skills/<skill-name>/
  SKILL.md
  kbtool                  # recommended entrypoint (POSIX wrapper)
  kbtool.cmd              # Windows wrapper
  kbtool.sha1             # stable hash of python sources
  kb.sqlite
  bin/
    <platform>/           # optional per-platform binary build (PyInstaller)
      kbtool(.exe)
      kbtool.sha1         # copy of kbtool.sha1 for freshness check
  scripts/
    kbtool.py             # python entrypoint (deterministic)
    kbtool_lib/           # implementation modules (db/search/bundle/hooks/skill-json…)
    reindex.py            # TSV-only reindex helper (fallback)
  catalog/
    categories.md
    categories/
  indexes/
    headings/            # sharded TSV title→path
    kw/                  # sharded TSV keyword→path (fallback only)
  references/
    <doc_id>/
      metadata.md
      toc.md
      chapters/
      sections/
```

## Robustness Rules (Do Not Skip)

- Prefer “structure-first”: if the document has headings, preserve them; don’t “chunk by size” unless necessary.
- Prefer deterministic search→bundle: run `./kbtool bundle --query "..." --out bundle.md`, then answer from `bundle.md`.
- Prefer the controlled retrieval path: let `bundle` perform 标题/正文/术语 三路召回 and inspect `## 补查记录` when a one-round expansion was triggered.
- Prefer “path-direct” only when user specifies chapter/section: open `references/<doc_id>/toc.md`, then open the target file.
- Treat `indexes/*` as fallback only; never load a whole large index file if a smaller shard or TOC suffices.

## Dependency Model (Cross-Platform)

- **Required:** `python3`
- **PDF (readable)**: uses `pdftotext` if available; otherwise instruct user to convert PDF → TXT/MD first.
- **DOCX:** uses a built-in OOXML extractor (no third-party Python deps); if extraction fails, instruct user to convert DOCX → MD/TXT.

## Pressure Scenarios (Self-Test)

- Missing dependencies: build from PDF on a machine without `pdftotext` (should fail with actionable instructions).
- Mixed inputs: build from `.md` + `.txt` + `.docx` in one run (should succeed).
- Rebuild safety: output skill folder already exists (should refuse unless `--force` is set).
- Version roll-forward: editing `references/` and running `python3 scripts/kbtool.py reindex` should create a shadow DB, validate it, then atomically switch.

## Common Mistakes

- Feeding a scanned PDF: this tool only supports *readable* PDFs; OCR first, or convert to TXT/MD.
- Assuming indexes are “the knowledge”: answers must cite `references/` files actually read; indexes are lookup only.
- Letting the model load huge files: always start from path-direct or per-doc `toc.md`, not `indexes/*` shards.

## Red Flags (Stop and Fix)

- “I’ll just open the whole index, it’s easier” → split the question and use TOC/shards.
- “PDF import failed, so I’ll guess” → stop; convert PDF to text or install `pdftotext`.

登录后下载/点赞/收藏 ❤ 7 | ★ 0

pack-builder

目录结构

SKILL.md

举报内容

提示