baissarienterprises-llm-wiki 0.2.1

Knowledge base operating system — AI agents compile raw sources into persistent, cross-linked Markdown wikis with concurrency control, multi-provider LLM integration, FTS5 search, and community detection

LLM Wiki Monorepo

pip install baissarienterprises-llm-wiki

A knowledge base that builds itself. AI agents compile raw sources into persistent, cross-linked Markdown wikis. Instead of re-retrieving documents on every query (RAG), the system incrementally builds and maintains a living wiki from your sources. Knowledge compounds over time. No database. No API lock-in. One repo. Any agent. Any machine.

---

Features

• Two-Step Chain-of-Thought Ingest — LLM analyzes sources first, then generates structured wiki pages with SHA256 caching, streaming Stage 1 progress, and multi-provider LLM support (OpenAI, Anthropic, DeepSeek, or agent-native with zero API keys)
• Agent-Native LLM Provider — routes ingest calls through Hermes, Claude Code, or Codex models directly — no external API keys required when running inside an AI agent session
• Structured Output — Pydantic-typed FILE and REVIEW blocks via instructor — no fragile regex parsing. Retry with exponential backoff on transient failures. Token counting and cost estimation per operation
• Concurrency Control — per-page advisory locking, atomic writes (temp → fsync → rename), content-hash conflict detection (SHA256 in frontmatter), and three-tier conflict management with automatic cleanup
• 15-Pass Automated Lint — dead links, orphans, frontmatter validation, contradictions, source drift, unresolved conflicts, stale page detection when raw sources change
• Knowledge Graph Engine — Louvain community detection with full Blondel et al. modularity, 4-signal relevance model with precomputed adjacency, surprising connection discovery, and knowledge gap detection. Pure Python fallback included
• SQLite FTS5 Search — full-text search with SHA256 freshness detection, pre-build at startup, incremental updates, BM25 fallback. --rebuild flag for full reindex
• Inverted Entity Index — dual-map entity→pages + page→entities for O(1) link suggestion lookups. 4-signal scoring with automatic wikilink insertion (--apply)
• MCP Server — 10 stdio tools for programmatic wiki access. Direct Python sidecar with zero subprocess overhead. Integrates with Claude Desktop, Codex, Cursor, and any MCP-compatible client
• Community Verification Suite — NMI/ARI cross-validation across 5 seeds, statistical similarity metrics, modularity tolerance within 1% relative error
• Backup & Recovery — tar.gz snapshots with restore, integrity verification, and automatic pruning. One-command --auto for safe state
• 19 Domain Templates — research, codebase, finance, machine learning, cybersecurity, medicine, algorithmic trading, and more. Every template provides PURPOSE.md, SCHEMA.md → CLAUDE.md, and extra-dirs.json
• Deep Research — web search → fetch → ingest → synthesize. Multi-source compilation into structured wiki pages
• Chrome Web Clipper — one-click web page capture with Readability + Turndown, auto-trigger ingest after clip
• CI/CD Pipeline — pytest + vitest matrix across Python 3.10–3.12 and Node 18–22, coverage reporting, caching, benchmark artifacts

Quick Start

# Install from PyPI
pip install baissarienterprises-llm-wiki

# Or install from source
git clone https://github.com/JeanBaissari/llm-wiki-monorepo.git
cd llm-wiki-monorepo
bash install.sh

# Scaffold a wiki
llm-wiki scaffold ~/my-wiki "My Research" --template research

# Ingest a source (two-step agent loop)
llm-wiki ingest ~/my-wiki raw/articles/my-source.md

# Use agent-native provider (no API keys — inside Hermes/Claude Code/Codex)
llm-wiki ingest ~/my-wiki raw/articles/my-source.md --llm opencode

# Check quality
llm-wiki lint ~/my-wiki

# Clean up old conflicts automatically
llm-wiki lint ~/my-wiki --clean-conflicts

# Build search index
llm-wiki index ~/my-wiki

# Discover hidden connections
llm-wiki insights ~/my-wiki

# Health check
llm-wiki health ~/my-wiki

# Start MCP server
llm-wiki serve ~/my-wiki

Architecture

wiki/ directory  ← shared state (Markdown files)
     │
     ├── Agent Skill + Python Scripts   → 20+ scripts: scaffold, ingest, lint,
     │                                     discover, insights, backup, link-suggest,
     │                                     deep-research, audit, benchmark, sidecar,
     │                                     lock_wiki, atomic_write, content_hash,
     │                                     index_wiki, louvain, health_check, wiki_logging
     ├── LLM Provider Layer             → openai, anthropic, litellm, opencode
     │   (src/llm_wiki/ + skill/scripts/providers/)
     ├── MCP Server (stdio)             → programmatic access, 10 tools,
     │                                     direct sidecar (zero subprocess)
     ├── Graph Engine (Node.js)         → relevance model, Louvain, insights,
     │                                     graphology bridge, verification suite
     ├── Web Viewer + Obsidian Plugin   → human browsing + feedback
     ├── Browser Extension              → web clipping + auto-ingest
     └── templates/                     → 19 domain schemas

Packages

Package	Language	Purpose
skill/	Python + Markdown	Agent skill (8 operations) + 20+ scripts + 12 reference docs
src/llm_wiki/	Python	PyPI package — CLI, LLM providers, concurrency, search, graph insights
mcp-server/	TypeScript	MCP server — 10 tools, direct sidecar integration
graph-engine/	TypeScript	Knowledge graph — relevance, Louvain communities, insights, verification
templates/	Markdown + JSON	19 domain-specific project templates
tests/	Python + TypeScript	pytest (ingest, lint, concurrency, search, opencode) + vitest (graph, mcp)
web-viewer/	TypeScript	Preview server with search + graph insights panel
extension/	JavaScript	Chrome web clipper with auto-ingest
audit-shared/	TypeScript	Shared audit file format library
plugins/obsidian-audit/	TypeScript	Obsidian plugin — file feedback from vault

Templates (19 domains)

research codebase finance algorithmic-trading cybersecurity machine-learning prompt-engineering copywriting marketing design-systems architecture crypto commodities decompilers medicine developer-tools personal-growth reading business

Every template provides: PURPOSE.md (scope + goals), SCHEMA.md → CLAUDE.md (page types, conventions, frontmatter, cross-referencing, contradiction handling), extra-dirs.json (domain directories).

Documentation

File	What it covers
README.md	You are here
QUICKGUIDE.md	Every command with real examples
AGENTS.md	Architecture, conventions, build/test commands, Python Dependency Policy
CHANGELOG.md	Full version history — all features, changes, and breaking changes
INDEX.md	Complete file tree with descriptions
VERSIONING.md	Semantic versioning policy and release process
PURPOSE.md	Why this system exists
skill/references/	12 detailed reference guides including concurrency, observability, and ingest

Requirements

• Python 3.10+ — for all skill scripts and PyPI package
• Node.js 18+ — for MCP server, graph engine, web viewer
• npm — for TypeScript package management
• pip dependencies — openai, anthropic, litellm, instructor, tenacity, tiktoken, python-dotenv, pydantic, portalocker (auto-installed via pip install)

Credits

The foundational methodology comes from Andrej Karpathy's llm-wiki pattern — using LLMs to incrementally build and maintain a personal wiki from raw sources. This project is a production-grade implementation with concurrency control, multi-provider LLM support, FTS5 search, MCP integration, and community detection.

License

MIT