nautilus-trader MCP server
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
nautilus-trader MCP server
What this is
An MCP (Model Context Protocol) server that gives an LLM (Claude Code, Codex CLI, etc.) two ways to look things up in nautilus_trader while writing strategy code against it:
Code search (
index.py) -- structural search over the actual source (Python + Rust), by symbol name. Not fuzzy-text/embedding search: it finds the real class/function/struct and returns its docstring or full source.Doc search (
rags/) -- semantic (embedding-based) search over the project's markdown docs (concepts, guides, tutorials), for "how do I..." questions that don't map to a single symbol name.
Related MCP server: MCPyDoc
How it works
index.pyspawns two real language servers as subprocesses and talks LSP (JSON-RPC over stdio) to them directly --pylspfor the Python source undernautilus_trader/python/nautilus_trader,rust-analyzerfor the Rust source undernautilus_trader/crates.documentSymbolfinds top-level classes/functions/structs/impls;hovergets the docstring (Python only -- see Known limitations). Results are cached to.code_index_cache.jsonsince building it costs ~25s (mostly rust-analyzer over ~2600 files).rags/rag_build.pychunks every file underdocs/and builds adspy.retrievers.Embeddingsindex (Gemini embeddings), saved torags/(config.json+corpus_embeddings.npy) plus ashape.jsonthat records the original folder structure sorags/search.pycan filter to a subfolder.mcp_server.pywires both into 5 MCP tools:search_code,get_code_doc,get_code_source,search_docs,show_doc_keys.
Setup
git clone https://github.com/GwangPyo/NautilusTraderMCP.git
cd NautilusTraderMCP
cp .env.example .env # fill in GEMINI_API_KEY (and OPENAI/ANTHROPIC if you use load_model)
./install.sh # conda env "mcp" + deps, nautilus_trader clone, code index, doc indexinstall.sh is idempotent: re-running it skips the nautilus_trader clone, the
docs/ copy, and the (paid) doc-embedding build if they already exist. Set
ENV_NAME=<name> to use a different conda env name (used for testing so it
doesn't touch the real mcp env).
An env manager isn't required -- uv venv && uv pip install -e . works too;
install.sh just standardizes on conda for a reproducible one-command setup.
Register with a client
./add_claude.sh # claude mcp add
./add_codex.sh # codex mcp addBoth just point the client at <conda mcp env>/bin/python3 mcp_server.py over stdio.
Known limitations / TODO
.code_index_cache.jsonandrags/{config.json,corpus_embeddings.npy,shape.json}have no invalidation -- ifnautilus_trader/ordocs/change, you have to delete the cache files by hand and re-run to pick it up.No automated tests -- everything so far has been verified by hand (fresh conda env, fresh uv env, real MCP client over stdio).
nautilus_trader/is cloned frommain(unpinned) -- can drift over time; nothing currently checks it against a known-good commit/tag.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Versioned documentation registry and semantic search for AI tools and coding assistants.
Code intelligence for LLMs. Analyze, search, and retrieve code from any public git repository.
Ask a codebase what calls what: search, blast radius, paths between symbols, and diffs.
Code intelligence for coding agents: semantic, AST, graph, and full-text search. 279+ languages.
Related MCP Servers
AlicenseNot gradedqualityDmaintenanceProvides AI coding agents with structured access to indexed codebases via semantic search, symbol analysis, and file reading tools.13Server Side Public , v 1- AlicenseNot gradedqualityDmaintenanceProvides real-time access to Python package documentation, source code, and symbol search to prevent AI hallucinations.710 PyPI7MIT
- AlicenseNot gradedqualityDmaintenanceExposes type-aware code navigation and fast file search to AI agents via language servers, enabling definitions, references, symbols, and file lookup without reading entire codebases.MIT
- AlicenseAqualityDmaintenanceEnables LLMs to efficiently navigate large codebases by providing surgical access to specific code symbols via semantic search and call-graph queries.6MIT