Searchclaw

Instructions

Implementation Reference

• src/index.ts:79-84 (registration)

  Registration of the 'search_ai' tool using server.tool() with name, description, schema, and handler

  server.tool(
    "search_ai",
    "RAG-ready web search with context and sources. Primary tool for AI agents — returns structured context optimized for LLM consumption. Costs 2 credits.",
    { q: z.string().describe("Search query") },
    async ({ q }) => jsonResult(await apiGet("/search/ai", { q }))
  );

• src/index.ts:82-82 (schema)

  Input schema definition for 'search_ai' tool: requires a 'q' string parameter for the search query

  { q: z.string().describe("Search query") },

• src/index.ts:83-83 (handler)

  Handler function for 'search_ai' tool: calls apiGet('/search/ai', { q }) and wraps result with jsonResult

  async ({ q }) => jsonResult(await apiGet("/search/ai", { q }))

• src/index.ts:20-39 (helper)

  apiGet helper function that makes HTTP GET requests to the SearchClaw API with timeout and error handling

  async function apiGet(path: string, params?: Record<string, string>) {
    const url = new URL(`${API_BASE}${path}`);
    if (params) {
      for (const [key, value] of Object.entries(params)) {
        url.searchParams.set(key, value);
      }
    }
    const controller = new AbortController();
    const timeout = setTimeout(() => controller.abort(), 30000);
    try {
      const response = await fetch(url.toString(), { headers, signal: controller.signal });
      if (!response.ok) {
        const text = await response.text();
        throw new Error(`SearchClaw API error ${response.status}: ${text}`);
      }
      return response.json();
    } finally {
      clearTimeout(timeout);
    }
  }

• src/index.ts:61-63 (helper)

  jsonResult helper function that formats API responses as MCP content with JSON text

  function jsonResult(data: unknown) {
    return { content: [{ type: "text" as const, text: JSON.stringify(data, null, 2) }] };
  }