> For the complete documentation index, see [llms.txt](https://docs.archisketch.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.archisketch.com/guide/dev-en/api-sdk/api-integration/mcp-integration.md). # MCP Integration Guide This guide provides instructions on integrating the MCP server to directly search Archisketch API documentation within AI coding tools. The Archisketch API documentation provides an MCP (Model Context Protocol) server. By connecting it to AI coding tools (Claude Desktop, Cursor, VS Code, etc.), you can search the Archisketch API documentation using natural language and efficiently proceed with integration tasks with the help of AI. ## 1. What is MCP? MCP (Model Context Protocol) is a standard protocol for delivering external data to AI models. When you connect to the Archisketch MCP server, the AI can directly search the API documentation to provide accurate answers and code based on the latest specifications. There is no need to manually browse the documentation or copy and paste document content for the AI. ## 2. MCP Server Information | **Item** | **Details** | | ---------------- | -------------------------------------------------------------- | | MCP Endpoint | `https://archisketch-docs.gitbook.io/archisketch/~gitbook/mcp` | | Transport Method | Streamable HTTP | | Authentication | Not Required (Public Document) | | Access Scope | Archisketch API Documentation (Latest Public Version) | ## 3. Setup Method by AI Tool Depending on the MCP configuration method of the AI or development tool you use, add the following JSON configuration, or utilize the commands provided below. Once configured, the AI tool will connect to the MCP server to search documents and understand your integration-related questions. ```json { "mcpServers": { "archisketch-docs": { "command": "npx", "args": [ "-y", "mcp-remote", "https://archisketch-docs.gitbook.io/archisketch/~gitbook/mcp" ] } } } ``` ### **3.1 Claude Code** Run the following command to add the server: ```bash claude mcp add --transport http --scope user archisketch-docs https://archisketch-docs.gitbook.io/archisketch/~gitbook/mcp ``` Verify if the MCP server has been successfully added with the following command: ```bash claude mcp list ``` ### **3.2 Cursor / VS Code** Add the JSON configuration shown above to your configuration file (`mcp.json` or `.cursor/mcp.json`). ## 4. Connection Verification Once the setup is complete, try asking the AI tool a question about the Archisketch API. If the AI searches the API documentation and provides specific guidance on endpoints, parameters, and response formats, it is successfully connected. If the connection fails: * Check if the endpoint URL is accurate. * Check if `archisketch-docs.gitbook.io` is accessible from your network. * Check if you have restarted the AI tool. * Check if there are any JSON syntax errors. ## 5. Use Cases ### **5.1 Starting API Integration** > "I'm about to start integrating the Archisketch API for the first time. How do I get started?" ### **5.2 Checking Specific API Specifications** > "Tell me the required parameters for the Archisketch Product Registration API." ### **5.3 Writing Integration Code** > "Write a code snippet to issue an Archisketch API authentication token in JavaScript." The AI will search the authentication API specification in the documentation and generate code reflecting the exact endpoint and parameters. ### **5.4 Error Troubleshooting** > "I'm getting an INVALID\_REQUEST error from the product registration API. What is the cause?" ## 6. FAQ **Q. Can I integrate the API without using MCP?** A. Yes. MCP is an optional convenience feature for development via AI tools. You can still check specifications directly on the API documentation website and integrate as before. **Q. Does the MCP server require authentication?** A. No. The Archisketch API documentation is public, so you can access the MCP server without separate authentication. However, an `x-api-key` is required when making actual API calls. **Q. Which AI tools can I use it with?** A. Any AI tool that supports MCP can be used. You can connect it via Claude Desktop, Cursor, VS Code, Claude Code, and more. **Q. Can I call the API directly using MCP?** A. No. MCP only provides the ability to search API documentation. Actual API calls must be requested directly to the Archisketch API server (`https://openapi.archisketch.com`) just as before. **Q. Are the latest documents reflected?** A. Yes. The MCP server always provides the latest public version of the documentation. When the documentation is updated, it is immediately reflected in the search results through MCP. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.archisketch.com/guide/dev-en/api-sdk/api-integration/mcp-integration.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.