Project Brain MCP

An MCP server that gives AI agents a cited, searchable memory of a codebase. Point it at a local folder or a public git URL. Claude, Cursor or any MCP client can then ask questions about the code and get answers backed by exact path:start-end citations.

Claude Code asks why TimestampSigner.unsign looks the way it does, what breaks if it changes, and whether its docs are current. Project Brain answers with commits, callers, tests and doc findings, all cited. (Waiting time sped up.)

Why it's different

• Code-aware chunking. Python is split by function, class and method using the real syntax tree, Markdown by heading. Each chunk knows its symbol (TimestampSigner.unsign) and exact lines, so citations point at real code.

• Hybrid retrieval. SQLite FTS5 BM25 (with camelCase / snake_case identifier splitting) plus local embeddings, merged with reciprocal-rank fusion. Exact names and fuzzy questions both work.

• Local and free. Embeddings run on your machine (fastembed, ONNX). No API keys, and no code leaves your computer.

• Incremental. Re-indexing only re-embeds files whose content changed. On a 42-file repo, a re-index with no changes takes about 2 seconds.

• History-aware. explain_history follows a function's lines back through git (git log -L), shows who wrote the current lines, and links the pull requests behind each change. So agents can answer "why is it built this way?", not just "what does it do?".

• Impact analysis. impact_of_change lists everything that uses a function, directly and transitively, plus the tests that exercise it, before an agent changes anything.

• Doc-drift detection. find_doc_drift checks Markdown and reStructuredText docs against the code. It finds files, functions, settings and flags that don't exist; examples passing parameters a function doesn't take; imports that would fail; and sections last edited long before the code they describe changed. Each finding cites the doc section and suggests a fix ("Was it renamed to checkout_total?").

• Agent-friendly output. Structured tool results, read-only tool annotations, progress notifications while indexing, and clear error messages the model can act on.

Related MCP server: OmniDocs-RAG-CN

Tools

Also included: the resource brain://repos, and the prompt answer_with_citations.

Quick start

uv sync
uv run project-brain index https://github.com/pallets/itsdangerous
uv run project-brain search "how is the signature timestamp checked for expiry"

0.0313  src/itsdangerous/timed.py:72-158  method TimestampSigner.unsign
0.0311  docs/timed.rst:1-28  code
0.0301  src/itsdangerous/timed.py:170-228  class TimedSerializer

Ask why it looks like that. explain_history(symbol="TimestampSigner.unsign") returns its line-by-line history:

src/itsdangerous/timed.py:72-158
  3edfbbbc 2022-03-24 handle OverflowError in timed unsign on 32-bit systems
  177196d3 2022-03-09 catch OSError on Windows
  37f09970 2022-03-08 catching year overflow ValueError
  c30678d1 2019-05-31 don't allow timestamps from the future
  910a7a94 2020-04-17 datetimes are timezone-aware utc
  ...

Use it from Claude Code

claude mcp add project-brain -- uv run --directory /path/to/project-brain-mcp project-brain

Configuration

Doc drift, tested honestly

On a well-maintained project (itsdangerous), find_doc_drift reports no false alarms. It only reports three "worth a look" sections whose docs date from 2020 while the code they describe changed in 2024–2025. On docs with planted mistakes, it catches every one: a moved file, a renamed function (with the right suggestion), an unknown setting and flag, a keyword argument the function doesn't take, and an import that would fail. Changelogs are skipped, because they're supposed to mention removed things.

How it works

source ─► resolve (local dir / shallow https clone) ─► list files (.gitignore-aware, skips binaries and lockfiles)
       ─► chunk (ast / headings / line windows) ─► embed (fastembed) ─► SQLite: chunks + vectors + FTS5
query  ─► BM25 top 50 ┐
       ─► vector top 50 ┴─► reciprocal-rank fusion ─► filters ─► cited sources

Security

• Clones are blobless (--filter=blob:none): full history, with file contents fetched only when needed.

• Commit authors are shown by name only; email addresses are never stored or returned.

• Only https:// clone URLs are accepted. Git runs with GIT_ALLOW_PROTOCOL=https:file and no terminal prompts.

• read_lines resolves paths and refuses anything outside the indexed repository.

• Repo names are validated, so they can't escape the index directory.

• All query tools are marked read-only.

Development

uv run pytest     # 38 tests, offline (hash embedder, temporary git repos, in-process MCP client)

Roadmap

• Git history and PR discussions: explain_history, search_history

• Symbol references and transitive dependents: impact_of_change

• find_doc_drift: Markdown and reStructuredText, signatures, imports, settings, flags, staleness

• tree-sitter chunking for more languages

• Streamable HTTP deployment with auth, retrieval evals, and OpenTelemetry tracing