Implements Karpathy's LLM Wiki pattern as a production-grade plugin. The LLM
incrementally ingests sources into a persistent, interlinked Obsidian vault —
updating entity/concept/source pages, flagging contradictions, maintaining an
index and append-only log. Knowledge compounds instead of being re-derived by
RAG on every query.
Plugin contents (engineering/llm-wiki/):
- SKILL.md with `context: fork` frontmatter for skill chaining
- 3 sub-agents: wiki-ingestor, wiki-librarian, wiki-linter
- 5 slash commands: /wiki-init, /wiki-ingest, /wiki-query, /wiki-lint, /wiki-log
- 8 Python tools (stdlib only): init_vault, ingest_source, update_index,
append_log, wiki_search (BM25), lint_wiki, graph_analyzer, export_marp
- 8 reference docs: schema, page-formats, ingest/query/lint workflows,
obsidian-setup, cross-tool-setup, memex-principles
- Vault templates: CLAUDE.md, AGENTS.md, .cursorrules, index.md, log.md,
5 page templates (entity, concept, source, comparison, synthesis)
- Worked example vault on "LLM interpretability"
- .claude-plugin/plugin.json manifest
Cross-tool compatibility: the scripts are pure Python stdlib. Only the schema
loader changes per tool (CLAUDE.md for Claude Code, AGENTS.md for Codex CLI /
Cursor / Antigravity / OpenCode / Gemini CLI, .cursorrules for legacy Cursor).
init_vault.py --tool all installs all three.
Repo-level registration:
- Commands mirrored to top-level commands/ for repo-wide discovery
- Agents mirrored to agents/engineering/ as cs-wiki-{ingestor,librarian,linter}
- .claude-plugin/marketplace.json: new llm-wiki entry + version bump to v2.3.0
- CLAUDE.md updated: 234 skills, 313 Python tools, 432 refs, 28 agents, 27 commands
Also saved (deferred): craighewitt-mattpocock reimplementation plan at
documentation/implementation/ — 4-pod proposal for building better versions
of selected skills from thecraighewitt-skills and mattpocock-skills
collections. Not executed; awaiting user confirmation on scope.
End-to-end smoke test passed: init_vault → ingest → update_index → append_log
→ wiki_search → lint → graph_analyzer → export_marp all run against a fresh
vault with real pages, wikilinks, and frontmatter.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Clean up repository by excluding internal planning and implementation
documents that are not relevant for end users.
Changes:
- Added documentation/implementation/* to .gitignore
- Removed SKILLS_REFACTORING_PLAN.md from git tracking
- File remains locally for maintainer use
Excluded Documents:
- documentation/implementation/SKILLS_REFACTORING_PLAN.md (internal planning)
- Future implementation docs in documentation/implementation/
Kept Documents (User-Facing):
- All root .md files (README, CONTRIBUTING, CHANGELOG, etc.)
- documentation/PYTHON_TOOLS_AUDIT.md (transparency about tool quality)
- documentation/GIST_CONTENT.md (excluded but committed initially)
Rationale:
- Root files follow open source best practices
- Python tools audit provides transparency
- Implementation planning is internal-only
- Cleaner repository for users
- Maintains professional appearance
All user-facing documentation remains accessible and comprehensive.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Create detailed systematic refactoring plan for optimizing all 36 skills
based on Anthropic's official Agent Skills specification and examples.
Plan Details:
- Complete comparison analysis of Anthropic vs current skills (Grade: B+)
- 4-phase implementation over 4 weeks
- Integrated metadata enhancement throughout
- Pilot optimization of 3 representative skills
- Full rollout to remaining 33 skills
- Testing, validation, and documentation
Key Optimizations:
1. Add professional metadata (license, version, category) to all skills
2. Add keywords sections for better discovery
3. Reduce SKILL.md files from avg 300 to 150 lines (50% reduction)
4. Move detailed content to references/ (progressive disclosure)
5. Add allowed-tools for security and safety
6. Maintain all domain expertise (reorganize, don't delete)
Expected Benefits:
- Faster skill loading (50-70% reduction in SKILL.md size)
- Better Claude activation (clearer triggers)
- Enhanced discovery (keywords + better descriptions)
- Professional versioning and tracking
- Safer execution (tool restrictions)
Implementation Tools Included:
- Metadata generator scripts
- Line counter for tracking progress
- Reference link validator
- Test protocol and success criteria
Total effort: ~40 hours over 4 weeks
Expected ROI: Permanent improvement to skill activation and performance
File location: documentation/implementation/SKILLS_REFACTORING_PLAN.md
Per project documentation structure requirements.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
