Frequently Asked Questions about UBOS Asset Marketplace’s MCP Server for MCP Servers

Q: What is an MCP Server? A: MCP (Model Context Protocol) is an open protocol that standardizes how applications provide context to LLMs. An MCP Server acts as a bridge, allowing AI models to access and interact with external data sources and tools, in this case, the Backlog API.

Q: What is the purpose of the Backlog MCP Server? A: The Backlog MCP Server enables seamless integration between AI agents and the Backlog API. It allows AI tools to manage projects, issues, wiki pages, and more within Backlog through AI-powered interactions.

Q: What AI agents are compatible with the Backlog MCP Server? A: The server is compatible with various AI agents, including Claude Desktop, Cline, Cursor, and any other AI tool that supports the Model Context Protocol.

Q: How do I install the Backlog MCP Server? A: The easiest way to install the server is via Docker. Configuration details for Claude Desktop or Cline are provided in the installation instructions. Alternatively, you can perform a manual installation by cloning the repository and building the project.

Q: What is GraphQL-style field selection? A: GraphQL-style field selection allows you to specify which fields you want to retrieve from the Backlog API. This optimizes responses by reducing data transfer and processing time, especially for large projects or complex issues.

Q: How can I optimize the response size? A: Enable the OPTIMIZE_RESPONSE option to use GraphQL-style field selection. Additionally, the server has a token limiting feature to prevent exceeding token limits.

Q: What is token limiting? A: Token limiting prevents exceeding token limits by automatically truncating large responses. The default limit is 50,000 tokens, configurable via the MAX_TOKENS environment variable.

Q: Can I customize tool descriptions? A: Yes, you can override tool descriptions by creating a .backlog-mcp-serverrc.json file in your home directory or by using environment variables.

Q: What tools are available through the MCP Server? A: The server provides a wide range of tools for interacting with the Backlog API, including tools for managing projects, issues, wiki pages, Git repositories, pull requests, notifications, and watching lists.

Q: How do I list all my Backlog projects using the MCP Server? A: You can ask your AI agent: “Could you list all my Backlog projects?”

Q: Can I create a new issue using the MCP Server? A: Yes, you can use a command like: “Create a new bug issue in the PROJECT-KEY project with high priority titled "Fix login page error"”

Q: Where can I find more information about the UBOS platform? A: You can visit the UBOS website at https://ubos.tech to learn more about the full-stack AI Agent Development Platform and its capabilities.

Q: Where can I report issues or seek support for the Backlog MCP Server? A: Report any issues via the GitHub Issues page. Please note that this tool is provided under the MIT License without any warranty or official support.