WEEEK MCP Server
Enables AI coding agents to read and write WEEEK projects, boards, tasks, and comments directly, without context switching.
README
weeek-mcp-server
MCP (Model Context Protocol) server for the WEEEK task tracker. Gives AI coding agents (Claude Desktop, Cursor, Cline, and any other MCP client) direct read/write access to WEEEK projects, boards, tasks, and comments — no context switching.
Features
- 12 tools — 7 read (projects, boards, columns, tasks, comments) + 5 write (create/update/move/complete tasks, post comments)
- Read/write split — tools are grouped so MCP clients can auto-approve reads while gating writes
- Stdio transport — zero server infrastructure, runs via
npx - Token auth — single
WEEEK_API_TOKENenv var, never logged - Safe defaults — list tools paginate (default 20, max 50) so responses stay under the 25k token MCP limit
- Structured errors — API failures return
isError: truewith a human-readable message, the server never crashes
Installation
No installation needed — configure your MCP client to run it via npx.
Getting a WEEEK API Token
- Sign in to WEEEK.
- Open Workspace settings → API.
- Generate a personal API token.
- Treat it like a password — it grants full read/write access to your workspace. Rotate it if it leaks.
Configuration
Claude Desktop
Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"weeek": {
"command": "npx",
"args": ["-y", "weeek-mcp-server"],
"env": {
"WEEEK_API_TOKEN": "your-weeek-token-here"
}
}
}
}
Restart Claude Desktop.
Cursor
Edit ~/.cursor/mcp.json (or use Cursor Settings → MCP → Add new MCP server):
{
"mcpServers": {
"weeek": {
"command": "npx",
"args": ["-y", "weeek-mcp-server"],
"env": {
"WEEEK_API_TOKEN": "your-weeek-token-here"
}
}
}
}
Generic MCP client
Any MCP client that supports stdio transport can launch:
command: npx
args: ["-y", "weeek-mcp-server"]
env: { WEEEK_API_TOKEN: "<your token>" }
NVM Workaround (IMPORTANT for nvm users)
If you installed Node via nvm, GUI applications (Claude Desktop, Cursor) do not source your shell startup files, so npx is not on their PATH. You will see spawn npx ENOENT in the client logs.
Fix: use the absolute path to your nvm npx binary.
-
In your terminal, run:
which npxExample output:
/Users/you/.nvm/versions/node/v22.0.0/bin/npx -
Put that absolute path in your MCP client config as the
command:{ "mcpServers": { "weeek": { "command": "/Users/you/.nvm/versions/node/v22.0.0/bin/npx", "args": ["-y", "weeek-mcp-server"], "env": { "WEEEK_API_TOKEN": "your-weeek-token-here" } } } }
This is the single most common setup failure across all npx-based MCP servers. If you upgrade Node via nvm, update the path.
Tools
All tools are prefixed weeek_. Read tools are side-effect free and safe for auto-approve. Write tools mutate WEEEK state and should prompt for user confirmation.
Read tools
| Tool | Purpose |
|---|---|
weeek_list_projects |
List projects in the workspace. Use FIRST to discover project IDs. |
weeek_get_project |
Get a single project's full details by ID. |
weeek_list_boards |
List boards inside a project. |
weeek_list_board_columns |
List columns (statuses) inside a board. Required before moving tasks. |
weeek_list_tasks |
List tasks with filters (project, board, column, assignee, completion) and pagination. |
weeek_get_task |
Get full details of a single task by ID. |
weeek_list_task_comments |
List comments on a task. |
Write tools
| Tool | Purpose |
|---|---|
weeek_create_task |
Create a NEW task. Requires title + project_id. |
weeek_update_task |
Edit fields (title, description, priority, assignee, due date) of an existing task. |
weeek_move_task |
Move a task to a different board column (status change). |
weeek_complete_task |
Mark a task complete, or reopen a completed task. |
weeek_create_task_comment |
Post a comment on a task. |
Safety
- Read/write separation: read tools and write tools are registered in separate groups. Configure your MCP client to auto-approve reads while gating writes.
- No delete operations: v1 intentionally does not expose delete endpoints — too destructive for an AI agent.
- Pagination defaults: list tools default to 20 results (max 50) to stay under the 25,000 token MCP response cap.
- Token handling:
WEEEK_API_TOKENis read from theenvblock only. It is never logged, echoed, or included in error messages.
Troubleshooting
| Symptom | Fix |
|---|---|
spawn npx ENOENT in client logs |
You are using nvm — see the NVM Workaround section above. |
WEEEK_API_TOKEN environment variable is required |
The env block is missing or empty in your MCP client config. |
Invalid WEEEK_API_TOKEN |
Token is wrong, revoked, or expired — regenerate in WEEEK workspace settings. |
| Server disconnects immediately after starting | You are on Node < 20. Upgrade Node (nvm install 20) and update your config path. |
| Tool returns "Resource not found (404)" | The ID doesn't exist in the workspace — list the parent resource first (e.g., weeek_list_projects before weeek_get_project). |
Development
git clone <this-repo>
cd weeek-mcp-server
npm install
npm run build
npm test
Scripts:
npm run build— compile TypeScript todist/npm run dev— run from source viatsxnpm run lint— ESLint (enforces no-console rule for stdio safety)npm run typecheck—tsc --noEmitnpm test— vitest unit tests
To smoke-test the built binary:
npm run build
WEEEK_API_TOKEN=test node dist/index.js
# server blocks on stdin — press Ctrl+C to exit
Requirements
- Node.js >= 20.0.0
- A WEEEK account with API access
License
MIT — see LICENSE.
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.