Frequently Asked Questions (FAQ) about Unbundle OpenAPI MCP Server

Q: What is the Unbundle OpenAPI MCP Server? A: It’s a tool that splits large OpenAPI specification files into smaller, manageable parts or extracts specific endpoints for easier use with AI models and applications.

Q: What is MCP? A: MCP stands for Model Context Protocol. It’s a standard that allows applications to provide context to Large Language Models (LLMs).

Q: How do I install the Unbundle OpenAPI MCP Server? A: You can install it globally using npm install -g @auto-browse/unbundle-openapi-mcp or run it directly with npx @auto-browse/unbundle-openapi-mcp@latest.

Q: What are the prerequisites for using this server? A: You need Node.js (LTS version recommended) and npm (which comes with Node.js) installed on your system.

Q: How do I configure the server with my MCP client (e.g., VS Code)? A: Add the server configuration details to your client’s settings file (e.g., settings.json in VS Code) as shown in the documentation.

Q: What MCP tools are provided by the server? A: The server provides split_openapi for splitting OpenAPI files and extract_openapi_endpoints for extracting specific endpoints into a new file.

Q: What does the split_openapi tool do? A: It executes the redocly split command to unbundle an OpenAPI definition file into multiple smaller files based on its structure.

Q: What arguments does split_openapi require? A: It requires apiPath (the path to the input OpenAPI file) and outputDir (the directory where the split files should be saved).

Q: What does the extract_openapi_endpoints tool do? A: It extracts specific endpoints from a large OpenAPI file and creates a new file containing only those endpoints and their referenced components.

Q: What arguments does extract_openapi_endpoints require? A: It requires inputApiPath (path to the large OpenAPI file), endpointsToKeep (an array of endpoint paths to extract), and outputApiPath (path where the new file should be saved).

Q: Can I use this server with Claude Desktop? A: Yes, you can configure it in your claude_desktop_config.json file as described in the documentation.

Q: Does this server require an internet connection? A: It may require an internet connection for npx to fetch @redocly/cli if it’s not already cached.

Q: What is UBOS? A: UBOS is a Full-stack AI Agent Development Platform focused on bringing AI Agents to every business department. It helps orchestrate AI Agents, connect them with enterprise data, and build custom AI Agents.

Q: How does the Unbundle OpenAPI MCP Server integrate with UBOS? A: It integrates seamlessly within the UBOS Asset Marketplace, providing a tool to manage OpenAPI specifications for use within UBOS AI agent development workflows.