Skip to main content
Glama
codyde

MCP Firecrawl Server

by codyde

extract-data

Extract structured data from websites using custom schemas and prompts to convert web content into organized formats for analysis and processing.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
urlsYes
promptYes
schemaYes

Implementation Reference

  • The handler function for the "extract-data" tool. It dynamically creates a Zod schema, calls Firecrawl's extract API with the URLs, prompt, and schema, handles errors with Sentry, and returns the extracted data as JSON or error message.
    async ({ urls, prompt, schema }) => {
      return await Sentry.startSpan(
        { name: "extract-data" },
        async () => {
          try {
            // Add Sentry breadcrumb for debugging
            Sentry.addBreadcrumb({
              category: 'extract-data',
              message: `Extracting data from URLs`,
              data: { urlCount: urls.length, prompt },
              level: 'info'
            });
    
            // Create the Zod schema from the provided definition
            const zodSchema = createDynamicSchema(schema);
    
            // Extract data using Firecrawl
            const extractResponse = await firecrawl.extract(urls, {
              prompt: prompt,
              schema: zodSchema
            });
    
            if (!extractResponse.success) {
              // Capture error in Sentry
              Sentry.captureMessage(`Failed to extract data: ${extractResponse.error}`, 'error');
              return {
                content: [{ 
                  type: "text", 
                  text: `Failed to extract data: ${extractResponse.error}` 
                }],
                isError: true
              };
            }
    
            return {
              content: [{ 
                type: "text", 
                text: JSON.stringify(extractResponse.data, null, 2)
              }]
            };
          } catch (error) {
            // Capture exception in Sentry
            Sentry.captureException(error);
            return {
              content: [{ 
                type: "text", 
                text: `Error extracting data: ${error.message}` 
              }],
              isError: true
            };
          }
        }
      );
    }
  • Input schema (Zod) for the "extract-data" tool defining urls (array of valid URLs), prompt (string), and schema (record supporting string, boolean, number, array, object types).
    { 
      urls: z.array(z.string().url()),
      prompt: z.string(),
      schema: z.record(z.union([
        z.literal('string'),
        z.literal('boolean'),
        z.literal('number'),
        z.array(z.any()),
        z.record(z.any())
      ]))
    },
  • src/server.js:130-131 (registration)
    Registration of the "extract-data" tool using McpServer's tool() method with the specified name.
    server.tool(
      "extract-data",
  • Helper function used by the extract-data handler to recursively build a Zod schema from the user-provided schema definition supporting primitive types, arrays, and objects.
    function createDynamicSchema(schemaDefinition) {
      const schemaMap = {
        string: z.string(),
        boolean: z.boolean(),
        number: z.number(),
        array: (itemType) => z.array(createDynamicSchema(itemType)),
        object: (properties) => {
          const shape = {};
          for (const [key, type] of Object.entries(properties)) {
            shape[key] = createDynamicSchema(type);
          }
          return z.object(shape);
        }
      };
    
      if (typeof schemaDefinition === 'string') {
        return schemaMap[schemaDefinition];
      } else if (Array.isArray(schemaDefinition)) {
        return schemaMap.array(schemaDefinition[0]);
      } else if (typeof schemaDefinition === 'object') {
        return schemaMap.object(schemaDefinition);
      }
      
      throw new Error(`Unsupported schema type: ${typeof schemaDefinition}`);
    }
Behavior1/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Tool has no description.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness1/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Tool has no description.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness1/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Tool has no description.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters1/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Tool has no description.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose1/5

Does the description clearly state what the tool does and how it differs from similar tools?

Tool has no description.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines1/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Tool has no description.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/codyde/mcp-firecrawl-tool'

If you have feedback or need assistance with the MCP directory API, please join our Discord server