本文へ移動
cccskills
無料GitHub で公開

llm-wiki

Build and maintain a structured LLM-generated wiki for any codebase. Use when the user asks to analyze/understand/document a codebase, build a code wiki, create project documentation from source, or update an existing .llm-wiki. Triggers on phrases like "build wiki", "analyze this codebase", "document this project", "update wiki", "llm-wiki", or when entering an unfamiliar project that has no .llm-wiki yet.

インストール方法を見る

含まれるファイル(3)

  • SKILL.md6.4 KB
  • references/article-templates.md3.7 KB
  • references/wiki-schema.md4.2 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

LLM Wiki for Codebases

Build a persistent, interlinked markdown wiki that captures the architecture, modules, patterns, and APIs of a codebase. The wiki lives in .llm-wiki/ at the project root. Humans curate and direct; the LLM handles all bookkeeping.

Based on Andrej Karpathy's LLM Wiki pattern: raw sources are "compiled" into a structured wiki that compounds over time.

Mode Detection

Determine the mode based on current state:

  • No .llm-wiki/ directory exists -> Full Build mode
  • .llm-wiki/ exists -> Update mode (diff and refresh)

Full Build Workflow

Phase 1: Reconnaissance

  1. Read top-level files: README, package.json/Cargo.toml/go.mod/pyproject.toml/build.gradle etc.
  2. Run find or Glob to map the directory tree (ignore node_modules, .git, vendor, dist, build, pycache, .venv)
  3. Identify: language(s), framework(s), build system, entry point(s), test framework
  4. Count files per directory to gauge module boundaries
  5. Read CLAUDE.md / AGENTS.md / .cursor/rules if present - they contain valuable architectural context

Record findings in .llm-wiki/_schema.md (see references/wiki-schema.md for format).

Phase 2: Skeleton

Create the directory structure:

.llm-wiki/
  _schema.md          # Wiki conventions and project metadata
  _index.md           # Content-oriented catalog by category
  _log.md             # Chronological build/update log
  architecture/       # High-level design docs
  modules/            # Per-module deep dives
  concepts/           # Cross-cutting concepts (auth, caching, error handling...)
  apis/               # API surface docs (REST endpoints, CLI commands, exported functions)
  guides/             # How-to guides (setup, deployment, testing)

Phase 3: Core Articles

Write articles in priority order. See references/article-templates.md for templates.

Priority 1 - Architecture:

  • architecture/overview.md - System architecture, component diagram (ASCII), tech stack
  • architecture/data-flow.md - How data flows through the system
  • architecture/directory-structure.md - Annotated directory tree

Priority 2 - Modules:

  • One modules/<name>.md per major module/package/directory
  • Cover: purpose, key files, public interface, internal patterns, dependencies

Priority 3 - Concepts:

  • Cross-cutting concerns that span modules (auth, logging, error handling, state management, config)
  • One concepts/<name>.md per concept

Priority 4 - APIs:

  • External-facing API surfaces (REST routes, CLI commands, SDK exports)
  • One apis/<name>.md per API group

Priority 5 - Guides:

  • guides/setup.md - Dev environment setup
  • guides/testing.md - How to run and write tests
  • Other guides as relevant

Phase 4: Index and Cross-link

  1. Build _index.md - organized by category with one-line descriptions and links
  2. Ensure every article has a ## See Also section linking to related articles
  3. Add backlinks: if A references B, B should reference A

Phase 5: Lint

Run a health check over the wiki:

  • Broken internal links (references to non-existent .md files)
  • Orphan pages (not linked from _index.md or any other page)
  • Missing coverage (directories/modules with no corresponding article)
  • Stale references (mentions of files/functions that don't exist in codebase)
  • Inconsistent terminology

Fix issues found. Log the lint run in _log.md.

Update Workflow

When .llm-wiki/ already exists:

  1. Read _schema.md to understand project metadata and conventions
  2. Read _log.md to see last update timestamp
  3. Detect changes since last wiki build:
    • git log --since="<last_update>" --name-status if git available
    • Otherwise, compare directory tree against architecture/directory-structure.md
  4. Triage changes:
    • New files/directories -> create new articles or update existing ones
    • Modified files -> re-read and update affected articles
    • Deleted files -> remove references, mark articles for cleanup
    • Renamed/moved files -> update paths in all referencing articles
  5. Update affected articles - re-read source, rewrite sections as needed
  6. Update _index.md if new articles added or old ones removed
  7. Run lint (same as Phase 5 above)
  8. Append to _log.md with timestamp, summary of changes

Writing Guidelines

  • Be factual: describe what the code does, not what it should do. Cite file paths and line ranges.
  • Use code snippets: short inline examples from actual source, not invented ones.
  • Link aggressively: every mention of another module/concept should be a markdown link to its article.
  • Keep articles focused: one topic per article, 100-500 lines. Split if longer.
  • Frontmatter: every article starts with a YAML frontmatter block:
    ---
    title: Module Name
    updated: 2026-04-09
    sources:
      - src/module/index.ts
      - src/module/utils.ts
    ---
    
  • ASCII diagrams over external images - they live in version control and render anywhere.
  • Language: match the project's primary language. If the codebase comments are in English, write in English. If Chinese, write in Chinese.

Agent Coordination

For large codebases (>500 source files), consider dispatching parallel agents:

  • Agent per top-level module to write module articles concurrently
  • One agent for architecture overview after modules are documented
  • One agent for cross-linking and lint

Gitignore

Add .llm-wiki/ to .gitignore only if the user prefers it. By default, the wiki is intended to be committed alongside the code so the team benefits.

Key Principles (from Karpathy)

  1. The wiki is the LLM's domain - humans rarely edit it directly
  2. Knowledge compounds - each query and exploration enriches the wiki
  3. Index files are critical - _index.md enables the LLM to navigate efficiently
  4. Lint regularly - catch rot before it spreads
  5. Log everything - _log.md provides temporal context for future updates

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

A股分析研究助手,提供行情数据获取与技术面/基本面分析框架(仅供研究参考,不构成投资建议)。适用于:(1) 获取A股行情和历史数据,(2) 技术面分析(K线形态、MACD、KDJ、RSI、布林带等),(3) 基本面分析(财务指标、估值分析),(4) 板块热点追踪,(5) 选股策略筛选与量化因子分析,(6) 生成股市分析报告。当用户询问"帮我分析股票"、"今日选股"、"A股行情分析"、"技术分析"、"量化选股"时触发。不用于:预测明日涨跌或给出确定性买卖指令、代客决策、港股美股(数据源不同)、加密货币。

日本語の概要は準備中です。原文の説明を表示しています。

staruhub/ClaudeSkills7282026年8月13日 更新

AI咨询/销售的对话策略助手。当用户需要准备AI方案沟通、跟业务部门聊AI落地、写AI提案、应对客户异议、做AI培训破冰时使用。触发场景:"怎么跟老板聊AI"、"客户说AI不靠谱"、"准备一个AI方案汇报"、"帮我想想怎么推AI"、"业务部门不配合"、"AI项目怎么卖"、"demo之后怎么跟进"。也适用于AI咨询师、技术合伙人、CTO做内部AI推广。

日本語の概要は準備中です。原文の説明を表示しています。

staruhub/ClaudeSkills7282026年8月13日 更新

Windows C盘清理和磁盘空间管理。当用户说C盘满了、磁盘空间不足、清理临时文件/缓存/回收站/系统日志、查找大文件、分析磁盘占用时使用。仅适用于 Windows 环境。不用于:macOS/Linux 磁盘清理、卸载软件(引导用户走系统卸载)、清理用户个人文件(只报告位置,删除决定权在用户)。

日本語の概要は準備中です。原文の説明を表示しています。

staruhub/ClaudeSkills7282026年8月13日 更新

PPT 生产 Agent:理解场景 → 推荐风格 → 先出大纲 → 定义页面 → 双通路交付(PPT 内容稿 / 逐页视觉图 / 信息图组图)。当用户要做汇报、路演、培训课件、提案、咨询报告的演示文稿,或要把大纲/文章做成 slides、deck、PPT、信息图、小红书图文、可视化传播图时使用。支持指定风格或从风格库推荐,支持 image 模型逐页出视觉稿。前身为 notion-infographic(v2)并吸收 ppt-designer 的设计原则。不用于:pptx 文件的纯技术操作(拆分/合并/提取,用宿主 pptx skill)、演讲稿或文章本体的写作(用 wechat-article-writer)、需要先做研究的内容调研(先走 deep-research 再回来)、单张海报或 logo 设计。

日本語の概要は準備中です。原文の説明を表示しています。

staruhub/ClaudeSkills7282026年8月13日 更新

Use this skill when the user wants an evidence-based research memo, literature review, market/policy/technical landscape, or a multi-source decision brief with citations, trade-offs, and a clear conclusion. Best for tasks that need synthesis across multiple external sources, iterative follow-up research, or a reusable written artifact. Do not use for quick factual lookups, single-source summaries, simple Q&A, summarizing one document the user already provided, plan-only requests where the user explicitly defers the actual research, or when the user wants a short answer with no need for cited evidence (a short but evidence-backed conclusion is still a brief memo, which this skill covers). Chinese trigger examples: "帮我调研", "深度研究", "综述报告", "技术选型分析", "竞品研究", "政策分析".

日本語の概要は準備中です。原文の説明を表示しています。

staruhub/ClaudeSkills7282026年8月13日 更新

资深高考命题专家助手,提供专业的命题指导和评审服务。适用于创作高考试题、评审试题质量、分析试卷结构、了解命题趋势等场景。结合文档工具提取解压文件,使用网络搜索了解当年最新命题趋势,使用分析工具评估题目质量和试卷结构。涵盖"一核四层四翼"评价体系、题型规范、评分标准、命题流程等多个维度。不用于:大学/考研/中考命题(体系不同,仅可借鉴)、日常作业题编写、直接替考生解题。

日本語の概要は準備中です。原文の説明を表示しています。

staruhub/ClaudeSkills7282026年8月13日 更新

staruhub のスキルをすべて見る

このスキルの問題を報告する