duckbrain 0.3.1

DuckDB-backed MCP memory server for Obsidian vaults — structured search, read, and write access for AI coding agents.

DuckBrain

DuckDB-backed MCP memory server for Obsidian vaults. Gives AI coding agents read/write access to your personal wiki — structured pages, full-text search, automatic indexing.

Installation

Pick your agent:

• OpenCode — MCP server + session plugin (recommended)
• Claude Code — MCP server + CLAUDE.md + SessionStart hook
• Cursor — MCP server + rules + hooks
• Hermes — MCP server + AGENTS.md

---

OpenCode

Best experience — session plugin gives the AI automatic vault awareness.

pip install duckbrain

Add to opencode.json:

{
  "mcp": {
    "duckbrain": {
      "command": "uv",
      "args": ["run", "duckbrain"],
      "env": { "VAULT_PATH": "/path/to/your/vault" }
    }
  }
}

Download the session plugin (no repo clone needed):

mkdir -p ~/.config/opencode/plugins/
curl -o ~/.config/opencode/plugins/vault-context.ts \
  https://raw.githubusercontent.com/timhiebenthal/duckbrain/main/opencode/plugins/vault-context.ts

The plugin makes the AI aware of your vault topics and recent daily notes automatically — no manual tool calls needed. It also adds a learnings ritual and journaling rule so the AI saves session progress on its own.

Restart OpenCode.

---

Claude Code

pip install duckbrain

Add to .claude/settings.json (project) or ~/.claude/settings.json (global):

{
  "mcpServers": {
    "duckbrain": {
      "command": "uv",
      "args": ["run", "duckbrain"],
      "env": { "VAULT_PATH": "/path/to/your/vault" }
    }
  }
}

CLAUDE.md

Add to .claude/CLAUDE.md:

# DuckBrain vault

Call vault_info() at session start to discover vault topics.
Use vault_search() or vault_read() when the query matches vault content.
Use vault_context() to load daily notes and search in one call.
After non-trivial work, save learnings with vault_write().

SessionStart hook (optional — auto-context)

Prototype — based on Claude Code docs, not yet validated end-to-end. Scripts work, but hook → injection pipeline needs manual verification.

For automatic vault awareness without manual tool calls, add a SessionStart hook. Download the script (no repo clone needed):

mkdir -p ~/.claude/hooks/
curl -o ~/.claude/hooks/vault-context.sh \
  https://raw.githubusercontent.com/timhiebenthal/duckbrain/main/scripts/claude-vault-context.sh
chmod +x ~/.claude/hooks/vault-context.sh

Add to the same .claude/settings.json:

{
  "hooks": {
    "SessionStart": [
      {
        "matcher": "startup",
        "hooks": [
          {
            "type": "command",
            "command": "/full/path/to/.claude/hooks/vault-context.sh"
          }
        ]
      }
    ]
  }
}

The hook injects vault tags and recent daily notes into Claude's context at session start — no manual vault_info() needed.

SessionEnd hook (optional — auto-journal)

Auto-stamp session end in today's daily note:

curl -o ~/.claude/hooks/vault-journal.sh \
  https://raw.githubusercontent.com/timhiebenthal/duckbrain/main/scripts/claude-vault-journal.sh
chmod +x ~/.claude/hooks/vault-journal.sh

Add to .claude/settings.json:

{
  "hooks": {
    "SessionEnd": [
      {
        "hooks": [
          {
            "type": "command",
            "command": "/full/path/to/.claude/hooks/vault-journal.sh"
          }
        ]
      }
    ]
  }
}

Appends ## Session end — HH:MM to today's daily note when the session ends.

Restart Claude Code.

---

Cursor

pip install duckbrain

Add to .cursor/mcp.json:

{
  "mcpServers": {
    "duckbrain": {
      "command": "uv",
      "args": ["run", "duckbrain"],
      "env": { "VAULT_PATH": "/path/to/your/vault" }
    }
  }
}

Optionally add .cursor/rules/duckbrain.mdc with alwaysApply: true:

---
description: DuckBrain vault knowledge base integration
alwaysApply: true
---
# DuckBrain vault
Call vault_info() at session start to discover vault topics.
Use vault_search() when the query matches vault content.

SessionStart hook (prototype — context injection)

Prototype — Cursor's additional_context from sessionStart hooks has a confirmed bug (dropped due to timing). The env output works, but context injection does not yet. Track: Cursor forum

mkdir -p ~/.cursor/hooks/
curl -o ~/.cursor/hooks/vault-context.sh \
  https://raw.githubusercontent.com/timhiebenthal/duckbrain/main/scripts/cursor-vault-context.sh
chmod +x ~/.cursor/hooks/vault-context.sh

Add to ~/.cursor/hooks.json:

{
  "version": 1,
  "hooks": {
    "sessionStart": [
      { "command": "/full/path/to/.cursor/hooks/vault-context.sh" }
    ]
  }
}

SessionEnd hook (auto-journal)

curl -o ~/.cursor/hooks/vault-journal.sh \
  https://raw.githubusercontent.com/timhiebenthal/duckbrain/main/scripts/cursor-vault-journal.sh
chmod +x ~/.cursor/hooks/vault-journal.sh

Add to ~/.cursor/hooks.json:

{
  "version": 1,
  "hooks": {
    "sessionEnd": [
      { "command": "/full/path/to/.cursor/hooks/vault-journal.sh" }
    ]
  }
}

Appends ## Session end — HH:MM to today's daily note when a session ends.

Restart Cursor.

---

Hermes

Add to mcp.json:

{
  "mcpServers": {
    "duckbrain": {
      "command": "uv",
      "args": ["run", "duckbrain"],
      "env": { "VAULT_PATH": "/path/to/your/vault" }
    }
  }
}

Add to AGENTS.md:

# DuckBrain vault
Call vault_info() at session start to discover vault topics.
Use vault_search() when the query matches vault content.
After non-trivial work, save learnings with vault_write().

Restart Hermes.

---

Tools

Tool	What it does
vault_search	Full-text search over vault pages
vault_read	Read a page by title or filepath
vault_write	Create a page or append to today's daily note
vault_context	Load daily notes + keyword search in one call
vault_info	Vault stats: page counts, tags, last modified

---

Vault Schema

your-vault/
├── wiki/
│   ├── entities/       # people, orgs, tools
│   ├── concepts/       # ideas, frameworks
│   ├── sources/        # summaries of ingested content
│   ├── synthesis/      # cross-cutting analysis
│   ├── index.md        # page catalog (auto-updated)
│   ├── log.md          # write history (auto-updated)
│   └── tags.md         # topic index (auto-updated)
├── daily/              # daily notes (YYYY-MM-DD.md)
└── .env                # VAULT_PATH (optional)

Pages use YAML frontmatter:

---
title: Claude Mem
item-type: entity
tags: [ai, memory, mcp]
created: 2026-05-28
updated: 2026-05-28
---

---

Nerdy Details

Implementation internals — not needed for installation.

Architecture

┌──────────────────────────────────────────────────────────────┐
│                      AI Agent                                │
│  ┌──────────────────────────┐  ┌──────────────────────────┐  │
│  │ MCP Client (stdio)       │  │ Hooks / Plugins          │  │
│  │  vault_search,           │  │ (SessionStart, system    │  │
│  │  vault_read, vault_write │  │  transform — inject      │  │
│  │  vault_context, vault_info│  │  vault context into      │  │
│  │                          │  │  system prompt)          │  │
│  └──────────┬───────────────┘  └──────────┬───────────────┘  │
└─────────────│──────────────────────────────│──────────────────┘
              │ MCP stdio                    │ reads directly
              ▼                              ▼ from vault
┌──────────────────────────────┐  ┌──────────────────────────────┐
│  DuckBrain MCP Server        │  │ Side channel:                │
│  vault_info   ──► DuckDB FTS │  │  wiki/tags.md                │
│  vault_search ──► DuckDB FTS │  │  daily/YYYY-MM-DD.md         │
│  vault_read   ──► Filesystem │  │  wiki/log.md                 │
│  vault_write  ──► Filesystem │  │                              │
└──────────────┬───────────────┘  └──────────────────────────────┘
               │ reads/writes
               ▼
┌──────────────────────────────────────────────────────────────────────┐
│                     Your Obsidian Vault                              │
│  wiki/entities/  wiki/concepts/  wiki/sources/  wiki/synthesis/      │
│  daily/          wiki/index.md   wiki/log.md    wiki/tags.md         │
└──────────────────────────────────────────────────────────────────────┘

• Reads vault files directly — no index to sync, no watchers, no duplicate storage
• Searches via DuckDB FTS (BM25 ranking), rebuilt fresh from disk on every query
• Writes new pages with YAML frontmatter, auto-updating index, log, and tags

Inspirations

This project stands on the shoulders of several ideas and tools:

• Andrej Karpathy's LLM wiki pattern — the idea that a personal markdown wiki, co-maintained by humans and AI agents, compounds into a persistent knowledge base. The vault schema (entities, concepts, sources, synthesis, daily log) is directly inspired by this.
• DuckDB — the embedded analytical database that makes full-text search over flat files viable without a server, index sync, or persistent storage. The decision to use in-memory FTS instead of a vector database was a deliberate trade-off for simplicity.
• Obsidian — the local-first, markdown-native note-taking tool that treats your files as the truth. DuckBrain exists because Obsidian vaults deserve tooling that respects the filesystem.
• MemSearch and Open Brain (OB1) — early experiments in cross-tool agent memory that demonstrated the need for structured vault write-back while choosing different architectures. Their strengths and gaps directly informed DuckBrain's design.
• Agent Memory Systems (6-level taxonomy) — Simon Scrapes' comprehensive comparison of Claude Code memory approaches provided the framework for understanding where DuckBrain fits in the ecosystem (Level 6: cross-tool MCP with dedicated server).
• trellis-datamodel — the same author's data modeling tool whose CI/CD patterns (trusted PyPI publishing, version-diff release detection, Keep a Changelog) were borrowed for this project's repository readiness.
• mondayDB 3 — Solving HTAP for a Trillion-Table System — monday.com's engineering blog on their DuckDB-powered CQRS read serving layer at production scale. Proved that DuckDB in-process with per-tenant file isolation is a viable architecture — the same pattern DuckBrain applies at personal-wiki scale.

The core decision — build, don't integrate — came from a structured comparison of 7 existing tools. All failed on one requirement: vault schema-aware write-back. Rather than fork or extend, DuckBrain started from first principles: what's the simplest thing that gives agents structured read/write access to an Obsidian vault? The answer was DuckDB + MCP + ~500 lines of Python.

---

Building from Source

git clone https://github.com/timhiebenthal/duckbrain.git
cd duckbrain
uv sync
uv run duckbrain  # will hang waiting on stdio — that's correct