MCP-Smallest.ai

What is MCP-Smallest.ai?

MCP-Smallest.ai is a Model Context Protocol (MCP) server that acts as middleware between client applications and the Smallest.ai API, providing a standardized interface for interacting with Smallest.ai's knowledge base management system.

How to use MCP-Smallest.ai?

Clone the repository, install dependencies with bun install, create a .env file with your SMALLEST_AI_API_KEY, then start the server with bun run index.ts. A test client can be run with bun run test-client.ts.

Key features of MCP-Smallest.ai

• Standardized MCP protocol communication for knowledge base operations
• Three built-in tools: listKnowledgeBases, createKnowledgeBase, getKnowledgeBase
• Documentation resource available at docs://smallest.ai
• Comprehensive error handling for HTTP, API, and parameter validation errors
• Type-safe responses and parameter validation
• Environment variable-based API key security

Use cases of MCP-Smallest.ai

• Listing all existing knowledge bases from the Smallest.ai API
• Creating new knowledge bases with a name and description
• Retrieving details of a specific knowledge base by its ID

FAQ from MCP-Smallest.ai

What tools does MCP-Smallest.ai provide?

Three tools: listKnowledgeBases, createKnowledgeBase, and getKnowledgeBase.

What are the runtime requirements?

Node.js 18+ or the Bun runtime, plus TypeScript knowledge.

How is the API key managed?

The API key is stored in the SMALLEST_AI_API_KEY environment variable inside a .env file.

What response format does the server use?

All responses return a JSON object with a content array containing a type: "text" and text field holding stringified data.

Does the server support authentication and error handling?

Yes, it implements comprehensive error handling (HTTP, API, parameter validation) and authenticates all requests through the API key.