Search...Search plugins and themes...
⌘K
Sign in
  • Get started
  • Download
  • Pricing
  • Enterprise
  • Account
  • Obsidian
  • Overview
  • Sync
  • Publish
  • Canvas
  • Mobile
  • Web Clipper
  • CLI
  • Learn
  • Help
  • Developers
  • Changelog
  • About
  • Roadmap
  • Blog
  • Resources
  • System status
  • License overview
  • Terms of service
  • Privacy policy
  • Security
  • Community
  • Plugins
  • Themes
  • Discord
  • Forum / 中文论坛
  • Merch store
  • Brand guidelines
Follow us
DiscordTwitterBlueskyThreadsMastodonYouTubeGitHub
© 2026 Obsidian

Mentat

Mentat TeamMentat Team310 downloads

Your local agentic cognitive partner for Obsidian, managing structured knowledge, local command operations, and deep mathematical reasoning.

Add to Obsidian
  • Overview
  • Scorecard
  • Updates11

Your local agentic cognitive partner for Obsidian. Mentat is an AI agent that lives in your vault — it reads, writes, searches, researches, and summarizes your notes through conversation.

What you can do with it

  • 💬 Chat with your vault: "Find my notes about RAG from last week and summarize them"
  • 🔍 Deep research: Create structured research plans, search the web, fetch pages, and save findings
  • ✍️ Write and edit notes: "Write a technical summary of this paper to Research/"
  • 🛠️ Automate tasks: Bulk operations, frontmatter updates, file moves with link fixing
  • 🎨 Two UI themes: Classic bubble chat or terminal-style timeline — switch anytime
  • 🧩 Extensible: Extension system, skill framework, MCP support

AI Providers

Provider Chat + Tools Embedding Local
OpenAI (and compatible) ✅ ✅ ❌
Anthropic Claude ✅ ❌ ❌
Ollama ❌ (no tool support) ✅ ✅

Set up in Settings → Mentat → Add AI Provider.

Quick start

# Install
cd /path/to/vault/.obsidian/plugins
git clone https://github.com/Secbone/obsidian-mentat mentat
cd mentat && npm install && npm run build

# Enable in Obsidian: Settings → Community plugins → Mentat
# Configure: Settings → Mentat → Add AI Provider
# Open chat: Ctrl/Cmd+P → "Open AI Chat"

Development

npm run build        # Type check + bundle → dist/
npm run deploy:dev   # Copy to ~/Documents/obsidian/.obsidian/plugins/
npm run lint         # ESLint
npm test             # Vitest

Project structure

src/
├── agents/          # Agent runtime (BaseAgent, Compactor, events)
├── chat/            # Chat orchestration (session, query, subagents)
├── context/         # Message windowing, token estimation
├── extensions/      # Extension system (EventBus, ExtensionManager)
├── providers/       # AI providers (OpenAI, Anthropic, Ollama)
├── prompts/         # System prompts, skill prompts, vault context
├── settings/        # Plugin settings & settings UI
├── skills/          # Skill framework (registry, executor, MCP, strategies)
├── ui/              # Themed UI (ChatView, bubble theme, terminal theme)
└── types/           # Shared type definitions

In-depth docs

Doc What it covers
src/ui/themes/README.md Theme system architecture, creating new themes
src/ui/README.md UI system overview, CSS conventions, icon sizing
src/agents/README.md Agent system, events, streaming, tool execution
src/skills/README.md Skill framework internals, MCP integration, custom skills
skills/README.md Adding and documenting skills
prompts/README.md Prompt system, template variables
docs/theme-system-refactor.md Theme system refactoring history

Roadmap

  • Phase 1: Agent system — conversation, tool calling, multi-turn reasoning
  • Phase 2: Skill framework — 12 built-in skills, progressive/native/auto modes, MCP
  • Phase 3: Themed UI — BubbleTheme, TerminalTheme, smart scroll, terminal presets
  • Phase 4: Extension system v1 + automatic context compaction
  • Phase 4.5: EventBus architecture (AsyncGenerator → EventBus, AgentContext.confirmHandler removed)
  • Custom skills: user-defined skills via .obsidian/plugins/mentat/skills/ with implementation.js
  • Phase 5: Tree-structured sessions with branching and comparison
  • Phase 6: Extension system v2 — dynamic loading, permission sandbox
  • Phase 7: i18n, plugin marketplace, community extension registry

Built with

  • Obsidian plugin API
  • Anthropic Claude · OpenAI · Ollama
  • PI Agent design influence (minimal core, extensibility)
  • OpenCode design influence (event-driven agent, permissions)

License

MIT

HealthExcellent
ReviewCaution
About
Organize notes, auto-classify and tag content with AI, and receive smart suggestions for linking related notes. Chat with your knowledge base via retrieval-augmented generation, visualize notes in an interactive 3D knowledge graph, and run spaced-repetition reviews while routing tasks across OpenAI-compatible, Anthropic, or local Ollama models.
AIResearchAutomation
Details
Current version
0.3.9
Last updated
3 months ago
Created
8 months ago
Updates
11 releases
Downloads
310
Compatible with
Obsidian 1.8.0+
Platforms
Desktop, Mobile
License
MIT
Report bugRequest featureReport plugin
Sponsor
GitHub Sponsors
Author
Mentat TeamMentat Teamsecbone
GitHubsecbone
  1. Community
  2. Plugins
  3. AI
  4. Mentat

Related plugins

Karpathy LLM Wiki

Karpathy's LLM Wiki implementation plugin for Obsidian - turns notes and PDFs into a linked, LLM-powered knowledge base with entity pages, concept pages, graph-powered Q&A, and local-first privacy.

Khoj

An AI personal assistant for your digital brain.

Smart Lookup

Semantic search for your vault. Ask in natural language, find notes by meaning when exact words fail, preview matching notes, and turn forgotten ideas into links, context, and next steps.

Smart Templates

AI powered templates for generating structured content. Works with Local Models, Anthropic Claude, Gemini, OpenAI & more.

Homepage

Open a note, base, or workspace on startup, or set it for quick access later.

Linter

Format and style your notes. Linter can be used to format YAML tags, aliases, arrays, and metadata; footnotes; headings; spacing; math blocks; regular Markdown contents like list, italics, and bold styles; and more with the use of custom rule options.

BRAT

Easily install a beta version of a plugin for testing.

Claudian

Embeds Claude Code/Codex and other local Agents as AI collaborators in your vault.

Smart Connections

Find related notes and excerpts while writing. Your AI link building copilot displays relevant content in graph + list view. A local embedding model powers semantic search. Zero setup. No API key.

Self-hosted LiveSync

Sync vaults securely to self-hosted servers or WEBRTC.