1
Add it to Claude Code
claude mcp add crawl4ai -- uv run crawl4ai-mcpEnvironment Variables
Set these before running Crawl4AI.
VariableDescriptionRequired
OPENAI_API_KEYAPI key for OpenAI models used in structured extractionNoANTHROPIC_API_KEYAPI key for Anthropic models used in structured extractionNoLLM_PROVIDERThe LLM provider to use for extraction tasksNoLLM_API_KEYGeneral API key for the selected LLM providerNoAvailable Tools (5)
Once configured, Crawl4AI gives your AI agent access to:
crawl_urlCrawl a single webpage and return content as Markdown.urlword_count_thresholdbypass_cachemagiccrawl_multipleCrawl multiple URLs concurrently.urlsmax_concurrentword_count_thresholdextract_structuredExtract structured data from a webpage using LLM instructions.urlinstructionproviderapi_tokenget_screenshotCapture a screenshot of a webpage.urlfull_pageviewport_widthviewport_heightdeep_crawlPerform a deep crawl of a website.urlmax_depthmax_pagesstrategyTry It Out
After setup, try these prompts with your AI agent:
→Crawl the documentation page at https://docs.crawl4ai.com and summarize the key features in Markdown.
→Extract all product names and prices from this e-commerce page: https://example.com/products
→Take a full-page screenshot of https://example.com and save it for my review.
→Perform a deep crawl of https://example.com up to a depth of 2 pages to gather site structure information.
Alternative installation methods
Claude Code (stdio)
claude mcp add crawl4ai uv run --project /path/to/crawl4ai-mcp crawl4ai-mcpKeep this setup from going cold
Save the docs, env vars, and workflow around Crawl4AI in Conare so Claude Code, Codex, and Cursor remember it next time.