← All tools

Firecrawl MCP Server

MCP

MCP server for Firecrawl web scraping integration. Supports both cloud and self-hosted instances. Features include web scraping, search, batch processing, structured data extraction, and LLM-powered content analysis.

v3.8.0 MIT Tested 7 Feb 2026
5.3

Dimension scores

Security 0.0
Reliability 6.0
Agent usability 7.0
Compatibility 8.0
Code health 8.0

Compatibility

Framework Status Notes
Claude Code
OpenAI Agents SDK ~ SSE transport implementation uses custom endpoints (/:apiKey/v2/sse) which may require adapter work for OpenAI SDK's expected SSE format, Complex nested Zod schemas in extraction tools may need flattening for OpenAI function calling format
LangChain Async batch operations may require careful state management in LangChain's execution model

Reliability

Success rate

72%

Calls made

100

Avg latency

2500ms

P95 latency

8000ms

Failure modes

  • No try-catch blocks around Firecrawl API calls - will crash on network errors or API failures
  • Missing parameter validation for required fields (url, apiKey) - undefined values will cause runtime errors
  • No timeout handling for long-running scrape/crawl operations - clients may hang indefinitely
  • Empty string/null handling not implemented - will pass invalid data to API
  • Concurrent request handling unclear - may exhaust resources with parallel batch operations
  • Error responses from Firecrawl API not structured into parseable format - raw errors passed through
  • No graceful degradation for rate limits beyond what Firecrawl library provides
  • Missing connection/resource cleanup in error paths
  • Unicode and special character handling delegated to Firecrawl library - behavior uncertain
  • Batch operations lack progress tracking on failures - no partial success handling visible

Code health

License

MIT

Has tests

Yes

Has CI

No

Dependencies

5

Well-maintained TypeScript MCP server with strong documentation and testing setup. The project has excellent documentation including README (25KB), CHANGELOG, VERSIONING guide, and LICENSE. TypeScript is fully configured with strict mode. Jest testing infrastructure is present with mock setup for the Firecrawl SDK. ESLint and Prettier configs indicate code quality focus. Package is published to npm at version 3.8.0. Dependencies are minimal (5 total) and include the Firecrawl SDK, dotenv, fastmcp, TypeScript, and Zod for schema validation. The CHANGELOG shows active recent development (latest entry from 2025-03-18) with meaningful fixes and features. Missing CI/CD configuration files (no .github/workflows, .travis.yml, etc.) is the main gap - tests exist but no automated runner. Docker support is present with both development and service configurations. Overall a professionally maintained project with minor gaps in automation.