tavily-extract

Instructions

Implementation Reference

• src/index.ts:290-305 (handler)

  Core handler function for the 'tavily-extract' tool. Makes a POST request to the Tavily extract API endpoint using the provided parameters (urls, extract_depth, include_images), handles API errors, and returns the response data.

  async extract(params: any): Promise<TavilyResponse> {
    try {
      const response = await this.axiosInstance.post(this.baseURLs.extract, {
        ...params,
        api_key: API_KEY
      });
      return response.data;
    } catch (error: any) {
      if (error.response?.status === 401) {
        throw new Error('Invalid API key');
      } else if (error.response?.status === 429) {
        throw new Error('Usage limit exceeded');
      }
      throw error;
    }
  }

• src/index.ts:174-195 (schema)

  Input schema for the 'tavily-extract' tool, defining parameters: urls (required array of strings), extract_depth (enum basic/advanced, default basic), include_images (boolean, default false).

  inputSchema: {
    type: "object",
    properties: {
      urls: { 
        type: "array",
        items: { type: "string" },
        description: "List of URLs to extract content from"
      },
      extract_depth: { 
        type: "string",
        enum: ["basic","advanced"],
        description: "Depth of extraction - 'basic' or 'advanced', if usrls are linkedin use 'advanced' or if explicitly told to use advanced",
        default: "basic"
      },
      include_images: { 
        type: "boolean", 
        description: "Include a list of images extracted from the urls in the response",
        default: false,
      }
    },
    required: ["urls"]
  }

• src/index.ts:171-196 (registration)

  Registration of the 'tavily-extract' tool in the ListToolsRequestSchema handler, including name, description, and input schema.

  {
    name: "tavily-extract",
    description: "A powerful web content extraction tool that retrieves and processes raw content from specified URLs, ideal for data collection, content analysis, and research tasks.",
    inputSchema: {
      type: "object",
      properties: {
        urls: { 
          type: "array",
          items: { type: "string" },
          description: "List of URLs to extract content from"
        },
        extract_depth: { 
          type: "string",
          enum: ["basic","advanced"],
          description: "Depth of extraction - 'basic' or 'advanced', if usrls are linkedin use 'advanced' or if explicitly told to use advanced",
          default: "basic"
        },
        include_images: { 
          type: "boolean", 
          description: "Include a list of images extracted from the urls in the response",
          default: false,
        }
      },
      required: ["urls"]
    }
  },

• src/index.ts:223-229 (handler)

  Dispatch logic in the CallToolRequestSchema handler that invokes the extract method when 'tavily-extract' is called.

  case "tavily-extract":
    response = await this.extract({
      urls: args.urls,
      extract_depth: args.extract_depth,
      include_images: args.include_images
    });
    break;

• src/index.ts:18-35 (helper)

  TypeScript interface defining the structure of the Tavily API response used by the tavily-extract tool.

  interface TavilyResponse {
    // Response structure from Tavily API
    query: string;
    follow_up_questions?: Array<string>;
    answer?: string;
    images?: Array<string | {
      url: string;
      description?: string;
    }>;
    results: Array<{
      title: string;
      url: string;
      content: string;
      score: number;
      published_date?: string;
      raw_content?: string;
    }>;
  }