agent-obsidian-mcp
Local MCP servers for the Agent-Obsidian workflow, providing vault file I/O, todo management, and arXiv paper fetching with zero-dependency single-file Node.js scripts.
README
agent-obsidian-mcp
Local MCP servers for the Agent-Obsidian workflow. One repo · one version tag · one GitHub Release; each Release attaches independently installable .tgz assets (built with npm pack).
| Package | Tool surface | Purpose |
|---|---|---|
vault-workspace-mcp |
vault_search · vault_read · vault_write · todo_query · todo_mark · workspace_lock_acquire · workspace_lock_release |
Agent file I/O, todo detection/marking, and the vault-wide write mutex. |
paper-fetch-mcp |
paper_fulltext_fetch |
arXiv full text with LaTeX math (ar5iv HTML → pdftotext fallback). |
Both servers are zero-dependency single files (Node ≥ 18 built-ins only; paper-fetch-mcp additionally shells out to pdftotext for its PDF fallback).
Deliberate boundaries
These servers are I/O capability layers only. Verdicts, scoring, state transitions, and completion belong to the workflow's gate core, which is implemented on the host — never here.
- No delete tool exists. Deletion is user-only in the workflow, so the surface is absent by construction.
vault_searchrequires a folder scope — full-vault scans are not offered.todo_markwrites only[/](agent-processing-finished). The final[x]is user-only.- Paths are vault-relative; absolute paths,
..traversal, and.obsidian/writes are rejected. VAULT_PATHenv is required forvault-workspace-mcp— there is no fallback path.
Develop
npm run smoke # offline smoke suites for both servers
SMOKE_NETWORK=1 node tests/smoke-paper-fetch.mjs # + one real arXiv fetch
npm run pack:all # build dist/*.tgz release assets
npm run digest # sha256 for the release manifest
Release & install (consumed by the workflow's bootstrap)
Release assets are the npm pack tgz files plus their sha256 digests. The bootstrap installs a pinned asset — never a floating latest:
npm install --omit=dev --ignore-scripts <verified .tgz>
MCP registration example (any MCP host):
vault-workspace:
command: node
args: [<install path>/node_modules/vault-workspace-mcp/server.mjs]
env: { VAULT_PATH: /absolute/path/to/vault }
paper-fetch:
command: node
args: [<install path>/node_modules/paper-fetch-mcp/server.mjs]
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.