Crawl4AI Tester
crawl4ai.geeoh.dev
Checking…
Connection Settings
Webhook: When set, async jobs (LLM & Crawl jobs) will POST their result to this URL when complete. See README for how to set up a local receiver on your Mac.
Note: Settings are saved in your browser's localStorage.
Quick Crawl
Fetch a URL and get clean markdown. Synchronous — waits for result.
POST /crawl
Options
Request & Response
Configure and run to see results
json

                

            

            

                

                    
bash

                    

                

            

            

                

                    
javascript

                    

                

            

            

                

                    
🕷
Run the crawl to see the response here

                

            

        

    


    
    

        

            

                

                    
Blog Discovery

                    
LLM-powered structured extraction. Submits an async job, then polls.

                

                POST /llm/job
            

            

                

                    
                    
                

                

                    
                    

                        
                        
                        
                        
                    

                

                

                    
                    
                

                

                    
                    
                

                

                
            

        

        

            

                

                    
Request & Response

                    
LLM extraction of structured post listings

                

                

            

            
            

                Task ID
                —
                
                
            

            

            

                
                
                
                
                
            

            

                

                    
json — POST /llm/job

                    

                

            

            

                

                    
bash — submit + poll

                    

                

            

            

                

                    
javascript — async + poll

                    

                

            

            

                

                    
🔍
Run discovery to see the raw API response

                

            

            

                

                    
📰
Extracted posts will appear here as cards

                

            

        

    


    
    

        

            

                

                    
Article Scraper

                    
Full article extraction with structured LLM output. One LLM job per URL.

                

                POST /llm/job
            

            

                

                    
                    
                

                

                    
                    
                

                

                    
                    
                

                

                
            

        

        

            

                

                    
Request & Response

                    
Full article extraction via LLM

                

                

            

            

                Task ID
                —
                
                
            

            

            

                
                
                
                
            

            

                

                    
json — POST /llm/job

                    

                

            

            

                

                    
bash — submit + poll

                    

                

            

            

                

                    
javascript — async + poll

                    

                

            

            

                

                    
📄
Run the scraper to see extracted article data

                

            

        

    


    
    

        

            

                

                    
Deep Crawl

                    
Multi-level crawl (BFS/DFS/Best-First). Async — job is queued and polled.

                

                POST /crawl/job
            

            

                

                    
                    
                

                

                    
                    

                        
                        
                        
                    

                

                

                    

                        
                        
                    

                    

                        
                        
                    

                

                

                    

                        
                        
                    

                    

                        
                        
                    

                

                

                    
                    
                

                

                
            

        

        

            

                

                    
Request & Response

                    
Multi-level site crawl

                

                

            

            

                Task ID
                —
                
                
            

            

            

                
                
                
                
            

            

                

                    
json — POST /crawl/job

                    

                

            

            

                

                    
bash — submit + poll

                    

                

            

            

                

                    
javascript — async + poll

                    

                

            

            

                

                    
🕸
Start the deep crawl to see multi-page results

                

            

        

    


    
    

        

            

                

                    
Retrieve Job

                    
Fetch the result of a previously submitted async job by Task ID

                

                GET /llm/job/:id
            

            

                

                    
                    
                

                

                    
                    

                        
                        
                    

                

                

                    
                    
                

                

                
Recent Jobs

                

                    
No recent jobs yet — submit a job from another tab first.

                

                

                
            

        

        

            

                

                    
Result

                    
Enter a Task ID and fetch

                

                

            

            

            

                
                
                
            

            

                

                    
📥
Enter a task ID and click Fetch Result

                

            

            

                

                    
bash

                    

                

            

            

                

                    
✨
Rendered content will appear here after fetch

                

            

        

    


    
    

        

            

                

                    
Markdown Extract

                    
Fetch a URL and convert to clean markdown. Synchronous.

                

                POST /md
            

            

                

                    
                    
                

                

                    
                    

                        
                        
                        
                        
                    

                

                
                

                
            

        

        

            

                

                    
Request & Response

                    
Configure and run to see results

                

                

            

            

                
                
                
                
                
            

            

                

                    
json

                    

                

            

            

                

                    
bash

                    

                

            

            

                

                    
javascript

                    

                

            

            

                

                    
📝
Run to see the raw JSON response

                

            

            

                

                    
✨
Rendered markdown will appear here

                

            

        

    


    
    

        

            

                

                    
Sanitized HTML

                    
Fetch preprocessed/sanitized HTML for a URL (useful for schema building).

                

                POST /html
            

            

                

                    
                    
                

                

                
            

        

        

            

                

                    
Request & Response

                    
Configure and run to see results

                

                

            

            

                
                
                
                
            

            

                

                    
json

                    

                

            

            

                

                    
bash

                    

                

            

            

                

                    
javascript

                    

                

            

            

                

                    
🧾
Run to see the sanitized HTML

                

            

        

    


    
    

        

            

                

                    
Screenshot

                    
Capture a full-page PNG screenshot of a URL.

                

                POST /screenshot
            

            

                

                    
                    
                

                

                    

                        
                        
                    

                    

                        

                            
                            
                        

                    

                

                

                
            

        

        

            

                

                    
Request & Response

                    
Configure and run to see results

                

                

            

            

                
                
                
                
                
            

            

                

                    
json

                    

                

            

            

                

                    
bash

                    

                

            

            

                

                    
javascript

                    

                

            

            

                

                    
📸
Run to see the raw JSON response (base64 PNG)

                

            

            

                

                    
🖼
Rendered screenshot preview will appear here

                

            

        

    


    
    

        

            

                

                    
PDF Snapshot

                    
Render a URL to a PDF document.

                

                POST /pdf
            

            

                

                    
                    
                

                

                
            

        

        

            

                

                    
Request & Response

                    
Configure and run to see results

                

                

            

            

                
                
                
                
                
            

            

                

                    
json

                    

                

            

            

                

                    
bash

                    

                

            

            

                

                    
javascript

                    

                

            

            

                

                    
📕
Run to see the raw JSON response (base64 PDF)

                

            

            

                

                    
📄
Rendered PDF preview will appear here

                

            

        

    


    
    

        

            

                

                    
Ask (LLM Q&A)

                    
Synchronous single-question LLM Q&A over a page. Requires a configured LLM key on the server.

                

                GET /llm/{url}
            

            

                

                    
                    
                

                

                    
                    
                

                

                    
                    
                

                

                
            

        

        

            

                

                    
Request & Response

                    
Configure and run to see results

                

                

            

            

                
                
                
                
            

            

                

                    
💬
Ask a question to see the LLM's answer here. If no LLM key is configured on the server, this will show a clear error instead.

                

            

            

                

                    
bash

                    

                

            

            

                

                    
javascript

                    

                

            

            

                

                    
📦
Run to see the raw JSON response

                

            

        

    


    
    

        

            

                

                    
Execute JS

                    
Run one or more JS snippets in the page context and capture return values.

                

                POST /execute_js
            

            

                

                    
                    
                

                

                    
                    
                

                

                    Note: Each line becomes one script string in the scripts array. If the server returns 403, the operator needs to set CRAWL4AI_EXECUTE_JS_ENABLED=true.
                

                

                
            

        

        

            

                

                    
Request & Response

                    
Configure and run to see results

                

                

            

            

                
                
                
                
                
            

            

                

                    
json

                    

                

            

            

                

                    
bash

                    

                

            

            

                

                    
javascript

                    

                

            

            

                

                    
🧬
Run to see the full CrawlResult JSON

                

            

            

                

                    
⚡
js_execution_result and basic page info will appear here

                

            

        

    


    
    

        

            

                

                    
Server Utilities

                    
Health check, config schema dump, and links to the built-in UIs.

                

                GET /health · /schema
            

            

                

                    
                    
                

                

                
Built-in UIs

                

                    🧪 Playground →
                    📊 Dashboard →
                

                

                    Tip: /health is public (no token needed); /schema requires the Bearer token from Settings and returns the full default BrowserConfig / CrawlerRunConfig field reference.
                

            

        

        

            

                

                    
Result

                    
Run a check to see results

                

                

            

            

                
                
            

            

                

                    
🩺
Run a health check or schema dump to see the response

                

            

            

                

                    
bash