MCP server enabling AI agents to perform natural knowledge discovery and analysis across Obsidian vault
Most planning and project tracking work still happens through a UI a human drives. Obsidian MCP server moves it into the conversation instead. MCP server enabling AI agents to perform natural knowledge discovery and analysis across Obsidian vault.
An MCP (Model Context Protocol) server that enables AI agents to perform sophisticated knowledge discovery and analysis across your Obsidian vault through the Local REST API plugin.
The server publishes 14 tools. What each one is for:
query — Text or regex search across note content (optional)query_type — Search type: "text" (default) or "regex"search_in_path — Limit search to specific folder pathtitle_contains — Filter by text in note titles (string, array, or JSON string)title_match_mode — How to match multiple terms: "any" (OR) or "all" (AND)tag — Filter by tag (string, array, or JSON string - searches frontmatter and inline #tags)tag_match_mode — How to match multiple tags: "any" (OR) or "all" (AND)context_length — Amount of content to return (set high for full content)include_content — Boolean to retrieve complete content of all matching notespage_size — Results per pagemax_matches_per_file — Limit matches per notepath — Directory to browse (defaults to vault root)Installation goes through your MCP client rather than a global install: point it at obsidian-api-mcp-server on PyPI and it is fetched when the client starts. The copy-paste blocks for Claude Desktop, Claude Code and Cursor are further down this page.
Configuration is passed through the environment: OBSIDIAN_API_URL, OBSIDIAN_API_KEY. Treat anything key-shaped as a real credential — scope it to the minimum the server needs, and rotate it if it ever lands in a shared config.
https://localhost:27124) and API key if you've set oneAmong the planning and project tracking options, the useful question is rarely "what can it do" but "what does it cost you to run" — permissions, credentials, and how much of your context its toolset consumes. Obsidian's toolset — query, query_type, search_in_path and 11 more — is a fair guide to whether it matches your workflow. It is maintained by pmmvr; worth a glance at recent repository activity before you build anything load-bearing on it.
This entry was verified against Obsidian's own documentation before publication; SyncDev keeps the directory reviewed rather than auto-generated.
| Tool | What it does |
|---|---|
| query | Text or regex search across note content (optional) |
| query_type | Search type: "text" (default) or "regex" |
| search_in_path | Limit search to specific folder path |
| title_contains | Filter by text in note titles (string, array, or JSON string) |
| title_match_mode | How to match multiple terms: "any" (OR) or "all" (AND) |
| tag | Filter by tag (string, array, or JSON string - searches frontmatter and inline #tags) |
| tag_match_mode | How to match multiple tags: "any" (OR) or "all" (AND) |
| context_length | Amount of content to return (set high for full content) |
| include_content | Boolean to retrieve complete content of all matching notes |
| page_size | Results per page |
| max_matches_per_file | Limit matches per note |
| path | Directory to browse (defaults to vault root) |
| include_files | Boolean to include files (default: False, folders only for speed) |
| recursive | Boolean to browse all nested directories |
{
"mcpServers": {
"obsidian-api": {
"command": "uvx",
"args": ["obsidian-api-mcp-server"],
"env": {
"OBSIDIAN_API_URL": "your-value",
"OBSIDIAN_API_KEY": "your-value"
}
}
}
}Add to claude_desktop_config.json, then restart Claude Desktop.
https://localhost:27124) and API key if you've set one| Variable | Description | Required |
|---|---|---|
| OBSIDIAN_API_URL | Endpoint or connection string the server talks to. | Yes |
| OBSIDIAN_API_KEY | Credential the server authenticates with. | Yes |
Your workspace, on speaking terms with AI — search, read and write Notion pages and databases.
Issue tracking at the speed of conversation — Linear's official hosted server with OAuth and zero install.
Search and read your Drive — Docs, Sheets and files become context your assistant can actually use.
Your assistant inside the workspace — read channels, search history, post messages and tame the noise.
Metabase ships its own MCP endpoint — search your BI content, build and run queries, and save questions and dashboards without leaving the chat.
Stop letting your assistant hallucinate n8n node parameters — this server hands it the real schemas, templates and validation.