v1.2.1

FireCrawl MCP Server v1.2.1 - Binary Configuration Hotfix

Changes

• Fixed binary configuration in package.json to resolve npx execution issues
• Simplified package configuration for better CLI tool support
• Added proper file inclusions for npm package distribution
• Updated package scripts to ensure proper build during installation

v1.2.0

FireCrawl MCP Server v1.2.0 - Enhanced Scraping & Search

Major Features

• 🔍 New search tool (fire_crawl_search) for web search with content extraction
• 🔄 Automatic retries with exponential backoff for rate limits
• 📊 Credit usage monitoring for cloud API operations
• 🚀 Queue system for batch operations with parallel processing
• 🌐 Support for self-hosted FireCrawl instances

Improvements

• Enhanced error handling for HTTP errors including 404s
• Improved URL validation before scraping
• Optimized rate limiting with automatic backoff strategy
• Added comprehensive logging system
• Updated configuration with new retry and batch processing options

Configuration Updates

• New FIRE_CRAWL_API_URL environment variable for self-hosted instances
• Automatic fallback to cloud API when no custom URL is provided
• Enhanced rate limit configuration with retry options
• Added batch processing parameters for parallel operations

Bug Fixes

• Rate limit handling in batch operations
• Error response formatting
• Type definitions for response handlers
• Test suite mock responses
• Error handling for invalid search queries
• API configuration validation

Requirements

• Node.js >= 18.0.0
• NPM or Yarn package manager

Installation

v1.0.2 - Fix JSON Communication

🐛 Bug Fixes

• Fixed JSON parsing error in MCP communication
• Removed plain text console.log that was breaking the protocol

🔧 Technical Details

• All communication now properly follows MCP JSON format
• Server initialization remains unchanged
• Improved compatibility with Claude Desktop

📝 Notes

• This is a critical fix for Claude Desktop integration
• No changes to core functionality
• Server status is now properly handled through MCP protocol

v1.0.1 - Path and Startup Message Fix

🔧 Fixes

• Fixed binary path in package.json to correctly point to dist/src/index.js
• Fixed start script path to match the dist structure
• Added startup message for better server status visibility

🛠️ Technical Details

• Updated bin path in package.json from dist/index.js to dist/src/index.js
• Updated start script to use correct path
• Added "FireCrawl MCP Server running on stdio" startup message

📝 Notes

• No changes to core functionality
• Improves server startup reliability
• Better integration with Claude Desktop

🔍 Verification

You can verify the server is running by checking the Claude logs, which should now show:

v1.0.0 - Initial Release

Features

• FireCrawl MCP Server implementation with advanced web scraping capabilities
• Support for single URL and batch scraping operations
• Multiple output formats (markdown, HTML, screenshots)
• Smart rate limiting and error handling
• Content filtering with include/exclude tags

Tools

fire_crawl_scrape

• Single URL scraping with customizable options
• Support for JavaScript-rendered content
• Mobile/Desktop viewport options

fire_crawl_batch

• Batch processing of multiple URLs
• Asynchronous operation with job tracking
• Same options as single scrape

fire_crawl_status

• Check batch job status and results
• Progress tracking with completion status

Technical Details

• Built with TypeScript and Model Context Protocol SDK
• Comprehensive test coverage
• ESLint and Prettier integration
• GitHub Actions for CI/CD
• Node.js 20.x support