CrawlForge MCP API Reference

https://crawlforge.dev/api/v1

curl -H "X-API-Key: cf_test_YOUR_KEY" \
     https://crawlforge.dev/api/v1/tools/fetch_url

Get your API key from the dashboard.

{
  "url": "https://example.com",
  "timeout": 10000,
  "follow_redirects": true
}

{
  "success": true,
  "data": {
    // Tool-specific response data
  },
  "credits_used": 1,
  "credits_remaining": 999,
  "processing_time": 245
}

A response can also carry an optional top-level warnings array of strings, sitting alongside data and credits_used rather than inside data. It is present only when a request produced a warning: you disabled the robots.txt check with respect_robots: false — which is also recorded against your API key — or a multi-URL tool skipped a URL or did not fetch a source. If you never override the check and every URL you request is allowed, no response carries the field at all; its absence is the normal case, not a missing value.

The official SDKs, crawlforge-sdk on npm and crawlforge on PyPI, expose one typed method per tool that takes the same parameters as the REST endpoint. The TypeScript and Python tabs on every tool page use them.

// npm install crawlforge-sdk
import { CrawlForge } from 'crawlforge-sdk';

const client = new CrawlForge({ apiKey: process.env.CRAWLFORGE_API_KEY });

const result = await client.fetchUrl({ url: 'https://example.com' });
console.log('Credits remaining:', result.creditsRemaining);

The whole REST API is described by an OpenAPI 3.1 document at https://www.crawlforge.dev/openapi.json, generated from the same schemas that validate every request, so a code generator or an API client can be pointed straight at it.

curl -s https://www.crawlforge.dev/openapi.json | jq '.paths | keys'

{
  "error": {
    "code": "VALIDATION_ERROR",
    "message": "Invalid input parameters",
    "details": [
      {
        "field": "url",
        "message": "Invalid URL format"
      }
    ]
  }
}

See the FAQ for a complete list of error codes and solutions.