firecrawl-cli

What This Skill Does

The firecrawl-cli skill provides powerful web scraping, crawling, and search capabilities directly within your OpenClaw environment. It is designed to transform complex, JavaScript-heavy web content into clean, LLM-optimized Markdown. By handling the heavy lifting of DOM manipulation, rendering, and content extraction, this skill enables you to pull research, technical documentation, or real-time web data into your context window efficiently and reliably.

Installation

To integrate this skill, use the ClawHub manager: clawhub install openclaw/skills/skills/leonardogrig/firecrawl-skills. Once installed, ensure you are authenticated by running firecrawl login --browser. You can verify your status, remaining API credits, and concurrency limits by executing firecrawl --status. Please ensure you follow your environment's standard rules for handling API keys and authentication workflows.

Use Cases

• Research and Intelligence: Quickly gather information on competitors, new tech trends, or market data by combining web searches with deep scraping.
• Documentation Extraction: Automate the retrieval of documentation pages from external developer portals to build internal knowledge bases.
• Content Monitoring: Track updates on specific websites or research pages over time.
• Data Pipelines: Generate structured JSON data from web search results to feed into further analysis workflows.

Example Prompts

1. "Use firecrawl to search for the latest benchmarks on Llama 3 and save the results as a JSON file in the .firecrawl directory."
2. "Crawl the documentation for the React Server Components guide at [url] and convert it into clean markdown for me to read."
3. "Search for recent news about AI agents from the past week, limit the results to 5 items, and save the output as search-ai-week.json."

Tips & Limitations

• Output Management: Always use the -o flag to write results to the .firecrawl/ directory. This keeps your main workspace clean and prevents token overflow in your active chat session.
• Rate Limiting: Monitor your firecrawl --status output. Do not exceed the concurrency limit of 100 jobs to ensure consistent performance.
• Security: Always add .firecrawl/ to your .gitignore to prevent accidental commits of scraped data. Note that this skill requires network access to function, as it interacts directly with external web servers to fetch content.