solanaprox-mcp

Instructions

Implementation Reference

• src/index.ts:324-342 (handler)

  Handler for 'list_models' tool that fetches capabilities from the API and formats the list of available AI models with their providers and pricing information

  case "list_models": {
    const capabilities = await getCapabilities();
  
    const modelList = capabilities.models
      ?.map(
        (m: any) =>
          `• ${m.id}\n  Provider: ${m.provider}\n  Cost: $${m.pricing?.input_per_1m || "?"}/1M input tokens`
      )
      .join("\n\n");
  
    return {
      content: [
        {
          type: "text",
          text: `🤖 Available Models on SolanaProx\n\n${modelList || "Claude Sonnet 4, GPT-4 Turbo"}\n\nPay with SOL or USDC. Deposit at ${SOLANAPROX_URL}`,
        },
      ],
    };
  }

• src/index.ts:102-111 (schema)

  Tool schema definition for 'list_models' - defines the tool name, description, and input schema (no parameters required)

  {
    name: "list_models",
    description:
      "List all available AI models on SolanaProx with their pricing.",
    inputSchema: {
      type: "object",
      properties: {},
      required: [],
    },
  },

• src/index.ts:192-196 (helper)

  Helper function that fetches capabilities from the SolanaProx API, including available models and their pricing

  async function getCapabilities(): Promise<any> {
    const res = await fetch(`${SOLANAPROX_URL}/api/capabilities`);
    if (!res.ok) throw new Error("Failed to fetch capabilities");
    return res.json();
  }

• src/index.ts:245-247 (registration)

  Registration of the tools array (containing list_models) with the MCP server's ListToolsRequestSchema handler

  server.setRequestHandler(ListToolsRequestSchema, async () => {
    return { tools };
  });