@praveen030686/data-apis-mcp

Instructions

Implementation Reference

• src/index.ts:381-384 (handler)

  The async handler function that calls the web extract batch API.

  async ({ urls }) => {
    const data = await apiPost(`${WEB_EXTRACT_API}/api/v1/extract/batch`, { urls });
    return { content: [{ type: "text", text: JSON.stringify(data, null, 2) }] };
  }

• src/index.ts:370-385 (registration)

  Registration of the web_extract_batch tool using server.registerTool.

  server.registerTool(
    "web_extract_batch",
    {
      title: "Batch Extract Text",
      description: `Extract text from up to 5 URLs in a single request.
  Costs $0.08 USDC per request via x402 on Base.`,
      inputSchema: {
        urls: z.array(z.string().url()).min(1).max(5).describe("Array of URLs to extract (max 5)"),
      },
      annotations: { readOnlyHint: true, destructiveHint: false, idempotentHint: true, openWorldHint: true },
    },
    async ({ urls }) => {
      const data = await apiPost(`${WEB_EXTRACT_API}/api/v1/extract/batch`, { urls });
      return { content: [{ type: "text", text: JSON.stringify(data, null, 2) }] };
    }
  );

• src/index.ts:376-378 (schema)

  Input schema definition for the tool, restricting input to an array of 1-5 URLs.

  inputSchema: {
    urls: z.array(z.string().url()).min(1).max(5).describe("Array of URLs to extract (max 5)"),
  },