The knowledge base that makes AI agents smarter over time.
Instead of retrieving raw fragments every query (RAG), your agent compiles, refines, and interlinks knowledge — like a team wiki that writes itself.
Works with Claude Code, Cursor, Windsurf, and any MCP client. No LLM built in — your agent IS the intelligence.
npx @agent-wiki/mcp-server serve --wiki-path ./my-knowledgeAdd to your MCP client config (Claude Code, Cursor, Windsurf, Claude Desktop):
{
"mcpServers": {
"agent-wiki": {
"command": "npx",
"args": ["-y", "@agent-wiki/mcp-server", "serve", "--wiki-path", "/path/to/knowledge"]
}
}
}That's it. Your agent now has a persistent, structured knowledge base.
| RAG | agent-wiki | |
|---|---|---|
| Approach | Retrieve fragments at query time | Build and maintain compiled knowledge |
| Memory | Stateless — forgets after each query | Persistent — knowledge accumulates |
| Quality | Raw chunks, often noisy | Curated, structured, interlinked |
| Cost | Embedding + retrieval every query | One-time compilation, free reads |
| Contradictions | Invisible — buried in source docs | Flagged automatically by lint |
| Source tracking | Lost after retrieval | Full provenance chain (raw -> wiki) |
| Feature | Description |
|---|---|
| Immutable Sources | SHA-256 verified raw/ layer — write-once, tamper-proof, full provenance |
| Knowledge Compilation | Agent builds structured wiki pages from raw sources — not retrieve-and-forget |
| BM25 Search | Field-weighted scoring, synonym expansion, fuzzy matching, CJK tokenization — zero LLM |
| Auto-Classification | Zero-LLM heuristic assigns entity types and tags across 10 categories |
| Self-Checking Lint | Catches contradictions, broken links, orphan pages, stale content |
| Atlassian Import | One-command Confluence pages and Jira issues with full hierarchy |
| File Versioning | Auto-version same-name files, query latest, list all versions |
| Directory Import | Point to a folder — imports all files with optional glob filtering |
| Document Extraction | PDF (with per-page access), DOCX, XLSX (multi-tab), PPTX — text extracted automatically |
| 16 MCP Tools | Full CRUD + search + lint + health checks |
| Git-Native | Plain Markdown — diffable, blameable, revertable |
Three immutability layers, inspired by how compilers work:
| Layer | Mutability | Role |
|---|---|---|
| raw/ | Immutable | Source documents — write-once, SHA-256 verified |
| wiki/ | Mutable | Compiled knowledge — structured pages that improve over time |
| schemas/ | Reference | Entity templates — consistent structure across knowledge types |
- Raw is immutable — Source documents are write-once, SHA-256 verified. Ground truth never changes.
- Wiki is mutable — Compiled knowledge improves with every interaction.
- No LLM dependency — Zero API keys, zero cost per operation. Your agent IS the intelligence.
- Self-checking — Lint catches structural issues and flags potential contradictions.
- Knowledge compounds — Every write enriches the whole wiki. Synthesis creates higher-order understanding.
- Provenance matters — Every wiki claim traces back to raw sources.
- Git-native — Plain Markdown. Every change is diffable, blameable, and revertable.
Inspired by Andrej Karpathy's LLM Wiki concept — the idea that AI agents should compile and maintain knowledge, not just retrieve raw fragments. This project is an independent, full implementation of that vision.
MIT