Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Rework MCP Servercreate a new task to review the project proposal by Friday"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Rework MCP Server
A Model Context Protocol (MCP) server for integrating Rework platform with AI applications. This server allows AI agents to interact with Rework projects, tasks, workflows, and jobs through a standardized protocol.
š Status Update: v1.1.3 now available with complete project management and workflow support.
Setup
Get your Rework credentials:
API key from Rework platform settings
Workspace information from your Rework account
Configure the MCP server with your credentials
Use natural language to manage your projects and workflows!
Installation
Add this entry to your client's MCP settings JSON file:
Alternatively, you can run the server directly using Node:
You can use the DISABLED_TOOLS environment variable to disable specific tools. Provide a comma-separated list of tool names to disable (e.g., create_task,get_tasks).
Disable tools you don't need if you are experiencing issues with the number of tools or any context limitations.
Running with SSE Support
The server can be run in SSE (Server-Sent Events) mode by setting the following environment variables:
Or via command line:
Docker Deployment
Features
š Project Management | š Workflow Management |
⢠Create and manage tasks ⢠View task details ⢠List tasks in projects ⢠Add comments to tasks ⢠Organize tasks within projects ⢠View task status and progress | ⢠Create and manage jobs ⢠Track job status ⢠View job details ⢠List jobs by workflow ⢠Organize workflows ⢠Link jobs to workflows |
š„ User Management | ā” Integration Features |
⢠Find users by name or email ⢠List all available users ⢠Assign users to tasks ⢠View user information ⢠User authentication ⢠Manage user permissions | ⢠ID and name-based lookups ⢠Markdown content support ⢠Built-in error handling ⢠API rate limiting ⢠Validation of inputs ⢠Comprehensive API coverage |
Available Tools
Tool | Description | Required Parameters |
Project Management | ||
| Create a task in a project |
|
| Get tasks from a board |
|
| Get detailed information about a task |
|
| Get all projects and boards | Optional search query |
Workflow Management | ||
| Create a job in a workflow |
|
| Get jobs from a workflow |
|
| Get detailed information about a job |
|
| Get all available workflows | Optional search query |
User Management | ||
| Find users by name or email |
|
| Get all workspace members | None |
See the project documentation for optional parameters and advanced usage.
User Management
When creating tasks or jobs, you can assign users using the user_id parameter. Additionally, you need to specify the creator_username to indicate who created the task or job:
The user management tools help you find users by name or email and get their IDs for task assignment.
Error Handling
The Rework MCP server provides clear error messages for:
Missing required parameters
Invalid IDs or references
Resources not found
Authentication failures
Permission issues
API errors
Rate limiting
The LOG_LEVEL environment variable can be specified to control the verbosity of server logs. Valid values are trace, debug, info, warn, and error (default).
This can also be specified on the command line as: --env LOG_LEVEL=info.
Custom Fields Support
Both task and job creation support custom fields through the custom_fields parameter, which accepts an array of objects with id and value properties:
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
License
This project is licensed under the MIT License - see the LICENSE file for details.
Disclaimer
This software makes use of the Rework API. All trademarks and brand names are the property of their respective owners. This project is not officially associated with or endorsed by Rework.