- Add AGENTS.md with repo guidelines
- Add lightrag-mcp: FastMCP server exposing insert_documents() + query_documents()
to LLM agents via stdio transport, talks to LightRAG REST API
- Add scripts/patch-vllm-cpu.py for CPU inference patching
- Add .env.vllm for vLLM configuration
- Update flake.nix with expanded dev shell
- Update .env.lightrag
- Remove CLAUDE.md (replaced by AGENTS.md)
Use subshells for cd to lightrag/ so user remains at repo root.
Add CLAUDE.md with purpose, architecture, and known issues.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
qwen3-embedding:8b needs >60s on first load. Set EMBEDDING_TIMEOUT=300
(worker executes at 2x = 600s) and TIMEOUT=600 for LLM calls.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds pkgs.bun to lightrag devShell and builds the React frontend
on first shell entry if lightrag/api/webui doesn't exist yet.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
ollama python package is in the offline-llm extra, not api.
Always run uv sync (not just on first entry) so extra changes take effect.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Impure devShells using uv + Python 3.12 for each project.
LD_LIBRARY_PATH set for NixOS native wheel compatibility.
Neo4j managed via nix run .#neo4j-start / neo4j-stop.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
tomatocream