What is FlowMCP (Core)?
FlowMCP (Core) is a schema-based framework that bridges existing web APIs (REST or GraphQL) with AI systems by standardizing interaction using the Model Context Protocol (MCP). It abstracts complex APIs into structured schema definitions, enabling seamless AI-driven communication with external services.
How to use FlowMCP (Core)?
Install the FlowMCP package, import the schema and server SDK, then call FlowMCP.activateServerTools() to register tools on an MCP server instance. Connect the server via StdioServerTransport(). Schemas are defined in .mjs files exporting a const schema = {} object following the specification.
Key features of FlowMCP (Core)
- Schema-based integration with structured, AI-interpretable route definitions.
- Modifier system for pre-, post-, and execute-phase logic injection.
- Inline parameters with Zod validation (string, number, boolean, enum).
- Built-in test cases per route to verify real-world API interactions.
- Text-based output in clear, human-readable format.
- Schema generator tool (ChatGPT-based) for automatic schema creation.
Use cases of FlowMCP (Core)
- Connecting an AI assistant to blockchain data APIs (e.g., Chainlink, Etherscan).
- Exposing a collection of REST endpoints as MCP tools for LLM consumption.
- Rapidly prototyping MCP-compatible servers from existing API documentation.
- Validating API schemas with embedded tests before deployment.
コメント