brain-mcp

<div align="center">

<img src="docs/images/brain-banner.jpg" alt="The example vault drawn as a brain: every dot is a note, every orange line a wikilink from the Harbor Ledger project hub" width="100%" />

# brain-mcp

**Your Obsidian vault, as memory for every AI you use.**

An open-source [Model Context Protocol](https://modelcontextprotocol.io) server that serves your notes, from your own machine, to Claude and any MCP client.

[![CI](https://github.com/debashishthakur/brain-mcp/actions/workflows/ci.yml/badge.svg)](https://github.com/debashishthakur/brain-mcp/actions/workflows/ci.yml)
[![npm](https://img.shields.io/npm/v/debawho-brain-mcp?logo=npm&color=CB3837)](https://www.npmjs.com/package/debawho-brain-mcp)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
[![PRs welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](CONTRIBUTING.md)
[![Good first issues](https://img.shields.io/github/issues/debashishthakur/brain-mcp/good%20first%20issue?label=good%20first%20issues&color=7057ff)](https://github.com/debashishthakur/brain-mcp/issues?q=is%3Aissue+is%3Aopen+label%3A%22good+first+issue%22)
[![Research issues](https://img.shields.io/github/issues/debashishthakur/brain-mcp/research?label=research%20questions&color=5319e7)](https://github.com/debashishthakur/brain-mcp/issues?q=is%3Aissue+is%3Aopen+label%3Aresearch)

![Node.js 22+](https://img.shields.io/badge/Node.js-22%2B-339933?logo=nodedotjs&logoColor=white)
![TypeScript](https://img.shields.io/badge/TypeScript-strict-3178C6?logo=typescript&logoColor=white)
![MCP](https://img.shields.io/badge/MCP-server-000000?logo=modelcontextprotocol&logoColor=white)
![SQLite FTS5](https://img.shields.io/badge/SQLite-FTS5-003B57?logo=sqlite&logoColor=white)
![Obsidian](https://img.shields.io/badge/Obsidian-vault-7C3AED?logo=obsidian&logoColor=white)
![Local models](https://img.shields.io/badge/models-local%20ONNX-FFD21E?logo=huggingface&logoColor=black)
![OAuth 2.1](https://img.shields.io/badge/OAuth-2.1%20%2B%20PKCE-EB5424)
![Cloudflare Tunnel](https://img.shields.io/badge/Cloudflare-Tunnel-F38020?logo=cloudflare&logoColor=white)

[**Website**](https://2brain.debawho.xyz) · [**Quick start**](#quick-start) · [**How it works**](#how-it-works) · [**Contribute**](#contributing) · [**Research questions**](#open-research-questions)

</div>

---

## Why brain-mcp

Every new chat starts from zero. You explain who you are, what you are building and how you like to work, and tomorrow, in another app, you explain it again.

brain-mcp keeps that context in the Markdown notes you already own. Connect any MCP client and it learns who you are, how you work, what your projects are and what you touched this week. It can search, follow links and write back to the vault as you talk.

- **Your files, your machine.** Notes stay plain Markdown in a folder you control. Nothing is uploaded to a third party.
- **One memory, every client.** Claude Code, Claude Desktop, claude.ai, Cursor, Hermes Agent, your phone and anything else that speaks MCP read the same vault.
- **Hybrid retrieval, fully local.** Keyword search plus meaning search with a small embedding model, then a reranker, all on your CPU. It can also answer "the vault does not record this" instead of guessing.
- **Measured.** Every ranking change is scored by two bundled evals, keyword and hybrid side by side.
- **Private by default.** OAuth 2.1 with PKCE for remote access, read, write and private scopes, secret redaction and an audit log of every call.
- **Writes back.** `brain_remember` and `brain_capture` turn what a model learns into notes you can read and edit.
- **Small and readable.** About 3,800 lines of strict TypeScript. Easy to study, easy to extend.

## How it works

```mermaid
flowchart LR
    V["Obsidian vault<br/>plain Markdown"]

    subgraph S["brain-mcp: one Node process, one SQLite file"]
        I["Indexer<br/>sections, links,<br/>topics, projects"]
        DB[("SQLite<br/>FTS5 word index<br/>section embeddings<br/>link graph")]
        M["Local models on CPU<br/>bge-small embedder<br/>bge-reranker"]
        R["Hybrid retrieval"]
        T["13 MCP tools"]
        G["Scopes, redaction,<br/>audit log"]
        I --> DB --> R --> T --> G
        M -.-> DB
        M -.-> R
    end

    V -->|file watcher| I
    G -->|stdio| C1["Claude Code,<br/>Claude Desktop"]
    G -->|HTTP + bearer token| C2["Clients on<br/>your network"]
    G -->|OAuth 2.1 through<br/>Cloudflare Tunnel| C3["claude.ai, phone,<br/>any MCP client"]
    T -. "remember, capture, write,<br/>edit, move, delete<br/>(allowed folders only)" .-> V
```

1. **Index.** Every note is split into sections and indexed with its links, topics and project. Each section also gets an embedding (`bge-small-en-v1.5`) in the background. A file watcher keeps both current within a second of a save.
2. **Retrieve.** A question goes through five steps:
   - **Understand:** expand shorthands (`pg` → PostgreSQL) and fix typos against the vault's own words.
   - **Search four ways:** keywords per section (BM25), meaning (embeddings), exact dates, and note titles.
   - **Fuse** the four lists with reciprocal rank fusion.
   - **Rerank** the best candidates with a cross-encoder (`bge-reranker-base`). When keyword and meaning search agree on the top hit, only the top 3 are reranked.
   - **Decide:** if even the best passage scores below a floor, answer "nothing relevant"; otherwise return the strong hits, cited by note, with a coverage label (good, thin, none).
3. **Serve.** Over stdio beside your editor, over HTTP with a bearer token on your network, or behind an OAuth 2.1 login through Cloudflare Tunnel for the public internet.
4. **Remember and write.** `brain_remember` appends durable facts and refuses near duplicates. Four more tools write, edit, move and delete hand-written notes, inside the folders you allow.

The retrieval step from question to answer:

```mermaid
flowchart LR
    Q["Question"] --> U["Understand<br/>shorthands,<br/>typo fixes"]
    U --> K["Keywords<br/>per section, BM25"]
    U --> D["Meaning<br/>embeddings"]
    U --> DT["Exact dates"]
    U --> NT["Note titles<br/>BM25"]
    K & D & DT & NT --> F["Fuse<br/>reciprocal rank fusion"]
    F --> RR["Rerank top 16<br/>cross-encoder<br/>(top 3 when keywords<br/>and meaning agree)"]
    RR --> GT{"Best score<br/>above the floor?"}
    GT -->|no| X["Nothing relevant"]
    GT -->|yes| A["Cited passages<br/>coverage: good or thin"]
```

Everything runs on your machine: the models are downloaded once into `data/models`, and no note leaves the server. If the models are not ready yet, retrieval falls back to keywords, so the server never blocks. Set `BRAIN_MCP_HYBRID=0` to stay keyword-only.

> [!NOTE]
> **The architecture is intentionally simple right now, and ideas are welcome.** It is one process and one SQLite file, with a hybrid ranker built from fixed rules. Some of it is already tweakable: the `retrieval` block in `brain.config.json` (embedding and reranker models, how many candidates to rerank, the score blend, the "nothing relevant" floor, your own shorthands), context budgets under `context`, and fusion constants such as `RRF_K` and `TITLE_BONUS` in `src/vault/index.ts`. Much more could become configurable, such as pluggable retrievers and storage, graph strategies, and query rewriting. If you have an idea, [open an issue](https://github.com/debashishthakur/brain-mcp/issues/new/choose) or a research proposal, even before there is code.

<p align="center">
  <img src="docs/images/graph-tour.jpg" alt="The project site: the example vault as a rotating brain, with one note lit and its wikilinks drawn in orange" width="92%" />
  <br />
  <sub>The bundled example vault as a knowledge graph on the <a href="https://2brain.debawho.xyz">project site</a>. Drag to rotate, hover a node to read the note.</sub>
</p>

### Tools

| Tool | What it returns |
| --- | --- |
| `brain_identity` | Persona bundle: profile note, memory notes, facts captured with `brain_remember`, skills, project list and recent focus. The server instructions ask clients to call this first. Pass `topic` for a smaller bundle. |
| `brain_context` | The sections most relevant to a question, with source note ids and a coverage label, from the hybrid ranker, packed under a size budget. Says so explicitly when the vault does not record the answer. |
| `brain_search` | Hybrid search (keywords, meaning, reranker, spelling correction) with project, type, topic and date filters. Returns ranked notes with snippets. |
| `brain_read` | One note, or one section of it, with metadata, topics and resolved links. |
| `brain_project` | Briefing on a project: description, latest progress, decisions, recent changes and notes grouped by type. With no argument it lists projects. |
| `brain_graph` | Links out, backlinks grouped by project, and notes that share topics. |
| `brain_recent` | Notes modified in the last N days, newest first. |
| `brain_capture` | Writes a new note with graph-ready frontmatter into the capture folder. |
| `brain_remember` | Appends a fact to the memory file; it joins `brain_identity` straight away. Near duplicates are refused (token Jaccard ≥ 0.6 or containment ≥ 0.85), partial overlaps get a `Supersedes` line, and `force=true` overrides. |
| `brain_write` | Creates a note at a chosen path in a writable folder (`Notes/` and the capture folder by default), or replaces one with `overwrite=true`. A `hub` note groups every note whose `project` matches its title. |
| `brain_edit` | Changes part of a note in one of three ways: replace text that matches once, replace a section under a heading, or append. Keeps the file's line endings and bumps `modified`. |
| `brain_move` | Moves or renames a note within the writable folders. Never overwrites, and names any notes whose links stop resolving. |
| `brain_delete` | Moves a note into `.trash/`, which is never indexed. Nothing is erased. |

The four write tools need the