Upload sources — PDFs, articles, notes, office docs. Review them in a full document viewer.
Connect Claude — via MCP. It reads your sources, writes wiki pages, maintains cross-references and citations.
The wiki compounds — every source you add and every question you ask makes it richer. Knowledge is built up, not re-derived.
Three Layers
| Layer | Description |
|-------|-------------|
| Raw Sources | PDFs, articles, notes, transcripts. Your immutable source of truth. The LLM reads them but never modifies them. |
| The Wiki | LLM-generated markdown pages — summaries, entity pages, cross-references, mermaid diagrams, tables. The LLM owns this layer. You read it; the LLM writes it. |
| The Tools | Search, read, and write. Claude connects via MCP and orchestrates the rest. |
Core Operations
LLM Wiki ships an MCP server that Claude.ai connects to directly. Once connected, Claude has tools to search, read, write, and delete across your entire knowledge vault. All operations below happen through Claude — you talk to it, it maintains the wiki.
— Drop a source in. Claude reads it, writes a summary, updates entity and concept pages across the wiki, and flags anything that contradicts existing knowledge. A single source might touch 10-15 wiki pages.
Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.
183,101
llm
llm-wiki
mcp
mcp-server
rag
supabase
Ingest
Query — Ask complex questions against the compiled wiki. Knowledge is already synthesized — not re-derived from raw chunks each time. Good answers get filed back as new pages, so your explorations compound.
Lint — Run health checks. Find inconsistent data, stale claims, orphan pages, missing cross-references. Claude suggests new questions to investigate and new sources to look for.
The tedious part of maintaining a knowledge base is not the reading or the thinking — it's the bookkeeping. Updating cross-references, keeping summaries current, noting when new data contradicts old claims, maintaining consistency across dozens of pages.
Humans abandon personal wikis because the maintenance burden grows faster than the value. LLMs don't get bored, don't forget to update a cross-reference, and can touch 15 files in one pass. The wiki stays maintained because the cost of maintenance drops to near zero.
The human's job is to curate sources, direct the analysis, ask good questions, and think about what it all means. The LLM's job is everything else.
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
