mcp-backlog-server

You run a Backlog MCP Server to bridge Backlog data with the Model Context Protocol. It exposes Backlog resources like projects, issues, and wiki pages, and provides tools to interact with Backlog data, generate summaries, and analyze usage from an MCP client side.

How to use

Use an MCP client to query Backlog resources and run prompts that summarize or analyze data. You can access a Backlog project, view issues and wiki pages, and invoke prompts to generate summaries of recently viewed projects or analyze usage patterns. Specific tools let you fetch user info, spaces, projects, issues, and wiki content, as well as add comments and retrieve detailed information.

How to install

Prerequisites: you need a system with Node.js and npm installed. Ensure you have an API key for Backlog and know your Backlog space URL.

Install dependencies for the Backlog MCP server project.

npm install

Build the server to produce the runtime artifacts.

npm run build

For development with automatic rebuilds, start the watch mode.

npm run watch

Configure the MCP server in Claude Desktop. Use the following snippet to add the backlog MCP server with the required environment variables.

{
  "mcpServers": {
    "backlog_mcp": {
      "command": "/path/to/mcp-backlog-server/build/index.js",
      "env": {
        "BACKLOG_API_KEY": "your-api-key",
        "BACKLOG_SPACE_URL": "https://your-space.backlog.com"
      }
    }
  }
}

Additional configuration and notes

Environment variables from the source are used to authenticate with Backlog and point to your Backlog space. The MCP server exposes a stdio interface that runs the built script at the specified path. If you deploy this locally, ensure the path to build/index.js matches your environment.

Security and troubleshooting

Keep your Backlog API key secret and restrict access to the MCP server. If you encounter startup issues, verify that the build artifacts exist at the configured path and that the Backlog space URL is reachable. Use MCP Inspector for debugging if available in your environment.

Notes on usage patterns

You can call tools to fetch current user info, space details, and recent projects, then drill into project issues and wiki pages. Prompts can summarize recently viewed projects, analyze backlog usage, and summarize wiki pages across a project.

Tools overview

The server provides a suite of tools to interact with Backlog data and generate insights. These tools cover retrieving user and space information, listing recent projects, accessing and filtering project issues, retrieving and updating wiki pages, and summarizing or analyzing data via prompts.

Available tools

get_backlog_user

Fetches the current Backlog user information.

get_backlog_space

Retrieves information about the Backlog space.

list_recent_projects

Lists recently viewed Backlog projects with optional count and sort order.

get_project_issues

Retrieves issues for a given project with filters like status and assignee.

get_issue_detail

Gets detailed information for a specific issue.

get_issue_comments

Fetches the comments associated with an issue.

add_issue_comment

Adds a new comment to an issue.

get_issue_comment_count

Returns the number of comments on an issue.

get_issue_comment

Retrieves a specific comment's details.

get_wiki_page_list

Lists wiki pages for a project with optional keyword filtering.

get_wiki_page

Fetches details for a specific wiki page.

update_wiki_page

Updates the content of a wiki page.

summarize_projects

Generates a summary of recently viewed Backlog projects.

analyze_backlog_usage

Analyzes Backlog usage patterns across users, spaces, and projects.

summarize_wiki_pages

Produces a summary of a project's wiki pages.