productboard
A **complete** Model Context Protocol (MCP) server that provides comprehensive integration with the Productboard API. More than 40+ tools.
claude mcp add --transport stdio enreign-productboard-mcp node /path/to/productboard-mcp/dist/index.js \ --env LOG_LEVEL="info" \ --env RATE_LIMIT_GLOBAL="100" \ --env RATE_LIMIT_WINDOW_MS="60000" \ --env PRODUCTBOARD_API_TOKEN="your-token-here" \ --env PRODUCTBOARD_AUTH_TYPE="bearer" \ --env PRODUCTBOARD_API_TIMEOUT="10000" \ --env PRODUCTBOARD_API_BASE_URL="https://api.productboard.com/v2"
How to use
This MCP server provides a suite of tools to interact with Productboard API v2 through the MCP protocol. It exposes 21 tools across five resource groups: Features, Products, Notes, Objectives & Key Results, and Releases. Each tool maps to a Productboard API operation, such as listing features, creating a feature, managing products and their hierarchy, handling customer feedback notes, and overseeing releases and their timelines. With this MCP server, AI assistants can read workspace data, create and update items, and fetch structured responses from the Productboard API, enabling interactive planning and management workflows. To use it, run the server and point your MCP-compatible client (like Claude Desktop or Claude Code) at the server configuration. Provide a valid Productboard API token via environment variables so the server can authenticate to the Productboard API. The available tools are grouped as follows: Features (pb_feature_list, pb_feature_get, pb_feature_create, pb_feature_update, pb_feature_delete), Products (pb_product_list, pb_product_create, pb_product_hierarchy), Notes (pb_note_list, pb_note_create), Objectives & Key Results (pb_objective_list, pb_objective_create, pb_objective_update, pb_keyresult_list, pb_keyresult_create, pb_keyresult_update), and Releases (pb_release_list, pb_release_create, pb_release_update, pb_release_status_update, pb_release_timeline). When querying, you can filter, shape, and paginate results according to the Productboard API capabilities, and the MCP layer will normalize responses for smooth AI consumption.
How to install
Prerequisites:
- Node.js v18+ (LTS)
- Access to a Productboard workspace with a valid API token
Installation steps:
- Clone the repository
git clone https://github.com/Enreign/productboard-mcp.git
cd productboard-mcp
- Install dependencies
npm install
- Build the project (TypeScript to JavaScript)
npm run build
- Run the MCP server (example)
# Start the server (adjust port/path if needed)
node dist/index.js
- Configure your MCP client (Claude Desktop or Claude Code) to point to the server, supplying your Productboard API token via environment variables
{
"mcpServers": {
"productboard": {
"command": "node",
"args": ["/path/to/productboard-mcp/dist/index.js"],
"env": {
"PRODUCTBOARD_API_TOKEN": "your-token-here"
}
}
}
}
Additional notes
Environment variables:
- Required: PRODUCTBOARD_API_TOKEN (Bearer token from Productboard)
- Optional/common: PRODUCTBOARD_AUTH_TYPE (bearer or oauth2), PRODUCTBOARD_API_BASE_URL (default https://api.productboard.com/v2), PRODUCTBOARD_API_TIMEOUT (ms), LOG_LEVEL (debug|info|warn|error), RATE_LIMIT_GLOBAL, RATE_LIMIT_WINDOW_MS. If using OAuth2, configure the additional OAuth-related variables as described in the README: PRODUCTBOARD_OAUTH_CLIENT_ID, PRODUCTBOARD_OAUTH_CLIENT_SECRET, PRODUCTBOARD_OAUTH_REDIRECT_URI. Ensure you keep tokens secret and do not commit them to source control. Common issues include token expiration, incorrect base URL, or insufficient API permissions on the Productboard token. If you encounter rate limit errors, adjust RATE_LIMIT_GLOBAL or RATE_LIMIT_WINDOW_MS accordingly and implement exponential backoff in client requests if supported by your MCP client.
Related MCP Servers
mcp-agent
Build effective agents using Model Context Protocol and simple workflow patterns
nerve
The Simple Agent Development Kit.
Remote
A type-safe solution to remote MCP communication, enabling effortless integration for centralized management of Model Context.
neurolink
Universal AI Development Platform with MCP server integration, multi-provider support, and professional CLI. Build, test, and deploy AI applications with multiple ai providers.
codemesh
The Self-Improving MCP Server - Agents write code to orchestrate multiple MCP servers with intelligent TypeScript execution and auto-augmentation
mcp-frontend
Frontend for MCP (Model Context Protocol) Kit for Go - A Complete MCP solutions for ready to use