Frequently Asked Questions (FAQ) about UBOS MCP Meilisearch API Server

Q: What is the MCP Meilisearch API Server? A: The MCP Meilisearch API Server is an implementation of the Model Context Protocol (MCP) that allows AI models to interact with the Meilisearch search engine. It provides a bridge between AI agents and Meilisearch, enabling AI models to leverage Meilisearch’s search functionalities.

Q: What is the Model Context Protocol (MCP)? A: MCP is an open protocol that standardizes how applications provide context to Large Language Models (LLMs). It defines a method for applications to expose their functionalities as “tools” that LLMs can invoke, allowing AI agents to access data and perform actions.

Q: What is Meilisearch? A: Meilisearch is an open-source, fast, and relevant search engine. It provides a user-friendly experience for developers and end-users, making it easy to index and search data.

Q: What are the key features of the MCP Meilisearch API Server? A: The key features include MCP compliance, full access to Meilisearch APIs, multiple transport options (STDIO and StreamableHTTP), real-time communication, and a comprehensive toolset for managing indexes, documents, and search settings.

Q: What are some use cases for the MCP Meilisearch API Server? A: Use cases include enhanced customer support, intelligent knowledge management, personalized recommendations, data-driven decision-making, and building context-aware AI agents.

Q: What types of tools are available in the MCP Meilisearch API Server? A: The server provides tools in the following categories: System Tools, Index Tools, Document Tools, Search Tools, Settings Tools, Task Tools, and Vector Tools. Each category contains tools for specific functionalities within Meilisearch.

Q: What is UBOS and how does the MCP Meilisearch API Server integrate with it? A: UBOS is a full-stack AI Agent development platform that helps businesses orchestrate AI agents, connect them with enterprise data, build custom AI agents, and create Multi-Agent Systems. The MCP Meilisearch API Server integrates seamlessly with UBOS, enhancing the capabilities of AI agents by providing access to real-time data, improving decision-making, and enhancing user experience.

Q: What are the prerequisites for using the MCP Meilisearch API Server? A: The prerequisites include Node.js v20 or higher, a running Meilisearch instance (local or remote), and an API key for Meilisearch (if required).

Q: How do I install the MCP Meilisearch API Server? A: You can install the mcp-meilisearch package using npm, yarn, or pnpm with the command npm install mcp-meilisearch, yarn add mcp-meilisearch, or pnpm add mcp-meilisearch.

Q: How do I configure the MCP Meilisearch API Server? A: You need to configure the server with options such as the Meilisearch host, API key, transport type, HTTP port, and MCP endpoint. These options can be set during server initialization.

Q: How can I use the MCPClient to interact with the server? A: You can use the MCPClient class to connect to the server and call various tools. First, create an instance of the MCPClient, then connect to the server using client.connectToServer(), and finally call the desired tool using client.callTool().

Q: What is vector search and how does the MCP Meilisearch API Server support it? A: Vector search is a type of semantic search based on vector embeddings. The MCP Meilisearch API Server provides experimental vector search capabilities, allowing you to perform searches based on the semantic meaning of the data. You can configure embedders and perform vector searches using the provided vector tools.

Q: How do I enable vector search in Meilisearch using the MCP Meilisearch API Server? A: You can enable vector search using the enable-vector-search tool. Additionally, you can configure embedders using the update-embedders tool to specify the models used to generate vector embeddings.

Q: What transport types does the MCP Meilisearch API Server support? A: The server supports both STDIO and StreamableHTTP transports, offering flexibility in how the server communicates with clients.

Q: What are the default connection options for Meilisearch? A: The default Meilisearch connection options are meilisearchHost: “http://localhost:7700” and meilisearchApiKey: “”.

Q: What is the default HTTP port for the MCP Server? A: The default HTTP port for the MCP server is 4995.