gtasks-mcp
Enables read/write access to Google Tasks, including listing, searching, creating, updating, deleting, and clearing tasks, from any MCP-compatible client.
README
gtasks-mcp
A Model Context Protocol server for Google Tasks — list, search, create, update, delete, and clear tasks from Claude (or any MCP-compatible client).
Adapted from zcaceres/gtasks-mcp (MIT), rebuilt on plain Node/TypeScript instead of Bun, with credentials moved out of the repo entirely (see Security below) and a full test suite.
Setup
1. Create a Google Cloud OAuth client
- Go to the Google Cloud Console, create a project (or reuse one).
- Enable the Google Tasks API (APIs & Services → Library).
- Configure the OAuth consent screen (External is fine for personal use — you'll be the only test user).
- Create credentials → OAuth client ID → Application type Desktop app.
- Download the JSON file.
2. Install the OAuth client file
mkdir -p ~/.gtasks-mcp
mv ~/Downloads/client_secret_*.json ~/.gtasks-mcp/oauth-keys.json
On Windows: %USERPROFILE%\.gtasks-mcp\oauth-keys.json.
3. Build and authenticate
npm install
npm run build
npm run auth
This opens a browser window for the Google consent flow. On success, a credentials.json (your refresh/access token) is written next to oauth-keys.json in ~/.gtasks-mcp/.
4. Register the server with Claude Code
claude mcp add gtasks -- node /absolute/path/to/gtasks-mcp/dist/index.js
Tools exposed
| Tool | Description |
|---|---|
list |
List tasks (all lists, or one via taskListId) |
search |
Search tasks by title/notes |
list-tasklists |
List all task lists |
create |
Create a task (title required) |
update |
Update a task (id required) |
delete |
Delete a task (id required) |
clear |
Clear completed tasks from a list |
Security
- Credentials never live inside this repo.
oauth-keys.json(your Google Cloud OAuth client) andcredentials.json(your personal access/refresh token, generated bynpm run auth) are both stored under~/.gtasks-mcp/— the OS user-config area, not the project directory. There is no gitignored file inside the repo to accidentallygit add -f; the secrets aren't reachable from the repo's working tree at all. Seesrc/config.tsfor the exact paths. credentials.jsonis equivalent to a password to your Google Tasks — anyone with that file can read/write your tasks. Treat it accordingly (don't email it, don't paste it in a chat, don't back it up somewhere shared).- Dependencies: only
@google-cloud/local-auth,googleapis, and@modelcontextprotocol/sdk— all official (Google / Anthropic). No telemetry, no third-party network calls; every request goes to Google's own APIs. - Known, accepted transitive vulnerabilities (
npm audit, checked 2026-08-25): moderate-severity issues inesbuild/vite(dev-only tooling, not part of the runtime) anduuid(viagaxios/googleapis, no untrusted-buffer input path in this server's actual usage). Both require a breaking major-version bump upstream to clear; revisit at the next periodic dependency audit rather than force an untested upgrade.
Development
npm test # vitest
npm run lint # eslint (narrow scope — see eslint.config.js)
npm run typecheck
npm run build
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.