Skip to content
Dang Nguyen edited this page Apr 18, 2026 · 18 revisions

Steward Wiki

Steward is an autonomous AI agent for Obsidian, powered by Large Language Models (LLMs). Equipped with tools and skills, it can search, manage your vault, and handle specialized tasks like creating Bases or Canvas files. Designed with simplicity and an immersive AI experience in mind, Steward lets you create your own commands, skills, and workflows to automate your boring and repetitive tasks.

Pages

  • Get started — How to start a new conversation in the editor or in the chat
  • Read content — Read notes, images, and specific content blocks
  • Edit — Edit notes with various modes (replace, insert, table operations, pattern-based)
  • Update frontmatter — Update YAML frontmatter properties of notes
  • Search — Fast BM25-based search with AI-powered natural language queries
  • Skills — Extend Steward with domain-specific knowledge for specialized tasks
  • Guardrails — Restrict AI access to specific folders and files
  • MCP — Connect Model Context Protocol servers and use their tools in conversations
  • CLI — Local shell: transcript (/>) vs interactive (node-pty), installers, and interactive program list
  • User-defined commands — Create custom commands to automate workflows
  • Revert — Undo any action performed by Steward
  • Settings — Plugin settings and configuration

Quick start

Installation

From Obsidian Community Plugins:

  1. Download the plugin from the Obsidian Community Plugins browser
  2. Enable the plugin in your Obsidian settings
  3. Configure your API keys in the plugin settings

Manual installation:

  1. Download the latest release from the releases page
  2. Extract the zip file into your vault's .obsidian/plugins folder
  3. Enable the plugin in your Obsidian settings
  4. Configure your API keys in the plugin settings

First steps

  1. Click the "Open Steward chat" icon in the ribbon to open the chat interface
  2. Type after the / (slash + space) to interact with Steward
  3. Type / ? to see available commands

Folder structure

Steward creates the following folder structure in your vault:

Steward/
├── Commands/       # Stores user-defined command definitions
├── Conversations/  # Archives past conversations
├── MCP/            # MCP server definitions (one markdown note per server)
├── Docs/           # Fetched documents from the repo
├── Release notes/  # Release notes of Steward
├── Rules/          # Guardrails rules (one file per rule)
├── Skills/         # Agent skills for domain-specific knowledge
├── Trash/          # Stores deleted files
└── Steward chat.md # Current active conversation

Key features

  • Built-in search engine — BM25-based search with relevant scoring and typo tolerance, significantly faster than native Obsidian search
  • Agent Skills — Extend Steward with domain-specific knowledge, compatible with the Agent Skills specification
  • Interactive chat UI — Chat interfaces using the slash / that leverage Obsidian's editor and reading view
  • Privacy-focused — Most actions run locally using Obsidian API, keeping your data safe
  • Model flexibility — Use OpenAI, Gemini, DeepSeek, Ollama, and more
  • Model fallback — Automatic failover to alternative models when errors occur
  • Intent caching — Embeddings-based caching to reduce token usage for similar queries
  • Multi-language support — Use Steward in your preferred language

Clone this wiki locally