awork MCP server
The awork MCP server gives AI assistants access to the full awork API, including projects, tasks, time entries, users and companies, using the permissions of the signed-in user or API token.
Written by
Connect awork
Claude Code, then run /mcp, select awork and choose Authenticate
claude mcp add --transport http awork https://api.awork.com/api/v1/mcpVisual Studio Code, in mcp.json (VS Code asks for the awork API key when the server starts)
{"servers": {"awork": {"url": "https://api.awork.com/api/v1/mcp", "type": "http", "headers": {"Authorization": "Bearer ${input:awork-api-key}", "Content-Type": "application/json", "Accept": "application/json"}}}, "inputs": [{"type": "promptString", "id": "awork-api-key", "description": "awork API Key", "password": true}]}Claude custom connector or Codex (Streamable HTTP), named awork: use this URL
https://api.awork.com/api/v1/mcpIn ChatGPT, open Apps, search for awork and select Connect, then log in with your awork credentials. awork's help page warns that connecting with an API key gives the MCP server admin permissions, so prefer OAuth where the client supports it.
Or paste this into your coding agent: Connect the awork MCP server (endpoint https://api.awork.com/api/v1/mcp) to my coding agent using the steps on https://developers.awork.com/mcp-server.
What the awork MCP server does
The awork MCP server connects AI assistants to awork, a project management tool that awork says it built specifically for agencies, covering project work, resource planning, teamwork and time logging. awork says the server provides access to all capabilities of its public API, so an assistant can work with projects, tasks, time entries, users, companies, files, webhooks and more.
Instead of exposing a separate MCP tool for every API endpoint, the server gives the language model guidance to generate the API code a request needs. The server then runs that code with your authenticated awork access and returns the result. Depending on the request, that can read data or create, update or delete it.
Everything is limited by the permissions of the signed-in user or API token, and the server uses the same rate limits as the standard awork API. Clients that support OAuth ask you to log in with your awork account; for clients that do not, awork documents an API key sent as a bearer token.
When to use awork
- You want an assistant to look up or update projects, tasks and time entries in awork.
- You already use Claude Code, Codex, ChatGPT or VS Code and want awork inside it.
- You want your agent to work with the same data as the public awork API without writing integration code.
When to pick something else
- A fixed list of named tools: awork documents none, because the server generates API code per request.
- Strict read-only use: the server can also create, update and delete data, limited only by the user's permissions.
What awork needs
- An awork account
- An MCP client that supports OAuth, or an awork API key
What the awork MCP server costs
The MCP server page names no separate fee. awork offers a free trial and paid plans, and its pricing page does not say which plan includes MCP access.
Current prices: www.awork.com . We do not list prices because vendors change them often. Check the publisher's own page.
Questions people ask
Does the awork MCP server work with ChatGPT?
Yes. awork's docs say it is available under Apps in ChatGPT, where you select Connect and log in with your awork credentials. They also describe setup for Claude, Claude Code, Codex and Visual Studio Code.
Is the awork MCP server free?
awork's MCP page names no separate fee. awork's pricing page (checked ) offers a free trial and paid plans, without saying which plan includes MCP access.
Other MCP servers for notes, docs and team tools
- Atlassian: The Atlassian MCP server gives AI agents access to Jira, Confluence, Jira Service Management, Bitbucket, Compass and Loom on Atlassian Cloud: search, summarize, and create or update issues and pages.
- Linear: The Linear MCP server gives AI agents access to your Linear workspace through Linear's hosted endpoint: find, create and update issues, projects and comments, with a read-only URL when you only need lookups.
- MeisterTask: The MeisterTask MCP server, in public beta, lets Claude work with the projects, tasks, Notes and mind maps in a Meister account, such as creating projects from a brief, updating task statuses and drafting Notes.
- MindMeister: The MindMeister MCP server, in public beta, lets AI clients such as Claude create, expand, restructure, summarize and search the mind maps in a MindMeister account, within that account's permissions.