Skrape MCP Server

Instructions

Implementation Reference

• src/index.ts:70-104 (handler)

  The main handler for the 'get_markdown' tool. It destructures arguments, validates the URL, calls the Skrape.ai API via axios, handles the response as markdown or JSON, and manages errors.

  case "get_markdown": {
    const { url, returnJson = false, options = { renderJs: true } } = request.params.arguments || {};
  
    if (!url || typeof url !== "string") {
      throw new McpError(ErrorCode.InvalidParams, "URL is required and must be a string");
    }
  
    try {
      const response = await axios.post(
        "https://skrape.ai/api/markdown",
        { url, options },
        {
          headers: {
            "Authorization": `Bearer ${API_KEY}`,
            "Content-Type": returnJson ? "application/json" : "text/markdown"
          }
        }
      );
  
      return {
        content: [{
          type: "text",
          text: returnJson ? JSON.stringify(response.data, null, 2) : response.data
        }]
      };
    } catch (error) {
      if (axios.isAxiosError(error)) {
        throw new McpError(
          ErrorCode.InternalError,
          `Skrape API error: ${error.response?.data?.message || error.message}`
        );
      }
      throw error;
    }
  }

• src/index.ts:36-62 (schema)

  Input schema for the 'get_markdown' tool defining the expected parameters: url (required), returnJson, and options with renderJs.

  inputSchema: {
    type: "object",
    properties: {
      url: {
        type: "string",
        description: "URL of the webpage to scrape"
      },
      returnJson: {
        type: "boolean",
        description: "Whether to return JSON response (true) or raw markdown (false)",
        default: false
      },
      options: {
        type: "object",
        description: "Additional scraping options",
        properties: {
          renderJs: {
            type: "boolean",
            description: "Whether to render the JavaScript content of the website",
            default: true
          }
        },
        default: { renderJs: true }
      }
    },
    required: ["url"]
  }

• src/index.ts:33-63 (registration)

  Registration of the 'get_markdown' tool in the ListTools handler response, providing name, description, and schema.

  {
    name: "get_markdown",
    description: "Get markdown content from a webpage using skrape.ai",
    inputSchema: {
      type: "object",
      properties: {
        url: {
          type: "string",
          description: "URL of the webpage to scrape"
        },
        returnJson: {
          type: "boolean",
          description: "Whether to return JSON response (true) or raw markdown (false)",
          default: false
        },
        options: {
          type: "object",
          description: "Additional scraping options",
          properties: {
            renderJs: {
              type: "boolean",
              description: "Whether to render the JavaScript content of the website",
              default: true
            }
          },
          default: { renderJs: true }
        }
      },
      required: ["url"]
    }
  }