File size: 10,594 Bytes
88f3fce |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 |
"""
Crawl4AI Web Crawler Tool for OpenManus
This tool integrates Crawl4AI, a high-performance web crawler designed for LLMs and AI agents,
providing fast, precise, and AI-ready data extraction with clean Markdown generation.
"""
import asyncio
from typing import List, Union
from urllib.parse import urlparse
from app.logger import logger
from app.tool.base import BaseTool, ToolResult
class Crawl4aiTool(BaseTool):
"""
Web crawler tool powered by Crawl4AI.
Provides clean markdown extraction optimized for AI processing.
"""
name: str = "crawl4ai"
description: str = """Web crawler that extracts clean, AI-ready content from web pages.
Features:
- Extracts clean markdown content optimized for LLMs
- Handles JavaScript-heavy sites and dynamic content
- Supports multiple URLs in a single request
- Fast and reliable with built-in error handling
Perfect for content analysis, research, and feeding web content to AI models."""
parameters: dict = {
"type": "object",
"properties": {
"urls": {
"type": "array",
"items": {"type": "string"},
"description": "(required) List of URLs to crawl. Can be a single URL or multiple URLs.",
"minItems": 1,
},
"timeout": {
"type": "integer",
"description": "(optional) Timeout in seconds for each URL. Default is 30.",
"default": 30,
"minimum": 5,
"maximum": 120,
},
"bypass_cache": {
"type": "boolean",
"description": "(optional) Whether to bypass cache and fetch fresh content. Default is false.",
"default": False,
},
"word_count_threshold": {
"type": "integer",
"description": "(optional) Minimum word count for content blocks. Default is 10.",
"default": 10,
"minimum": 1,
},
},
"required": ["urls"],
}
async def execute(
self,
urls: Union[str, List[str]],
timeout: int = 30,
bypass_cache: bool = False,
word_count_threshold: int = 10,
) -> ToolResult:
"""
Execute web crawling for the specified URLs.
Args:
urls: Single URL string or list of URLs to crawl
timeout: Timeout in seconds for each URL
bypass_cache: Whether to bypass cache
word_count_threshold: Minimum word count for content blocks
Returns:
ToolResult with crawl results
"""
# Normalize URLs to list
if isinstance(urls, str):
url_list = [urls]
else:
url_list = urls
# Validate URLs
valid_urls = []
for url in url_list:
if self._is_valid_url(url):
valid_urls.append(url)
else:
logger.warning(f"Invalid URL skipped: {url}")
if not valid_urls:
return ToolResult(error="No valid URLs provided")
try:
# Import crawl4ai components
from crawl4ai import (
AsyncWebCrawler,
BrowserConfig,
CacheMode,
CrawlerRunConfig,
)
# Configure browser settings
browser_config = BrowserConfig(
headless=True,
verbose=False,
browser_type="chromium",
ignore_https_errors=True,
java_script_enabled=True,
)
# Configure crawler settings
run_config = CrawlerRunConfig(
cache_mode=CacheMode.BYPASS if bypass_cache else CacheMode.ENABLED,
word_count_threshold=word_count_threshold,
process_iframes=True,
remove_overlay_elements=True,
excluded_tags=["script", "style"],
page_timeout=timeout * 1000, # Convert to milliseconds
verbose=False,
wait_until="domcontentloaded",
)
results = []
successful_count = 0
failed_count = 0
# Process each URL
async with AsyncWebCrawler(config=browser_config) as crawler:
for url in valid_urls:
try:
logger.info(f"🕷️ Crawling URL: {url}")
start_time = asyncio.get_event_loop().time()
result = await crawler.arun(url=url, config=run_config)
end_time = asyncio.get_event_loop().time()
execution_time = end_time - start_time
if result.success:
# Count words in markdown
word_count = 0
if hasattr(result, "markdown") and result.markdown:
word_count = len(result.markdown.split())
# Count links
links_count = 0
if hasattr(result, "links") and result.links:
internal_links = result.links.get("internal", [])
external_links = result.links.get("external", [])
links_count = len(internal_links) + len(external_links)
# Count images
images_count = 0
if hasattr(result, "media") and result.media:
images = result.media.get("images", [])
images_count = len(images)
results.append(
{
"url": url,
"success": True,
"status_code": getattr(result, "status_code", 200),
"title": result.metadata.get("title")
if result.metadata
else None,
"markdown": result.markdown
if hasattr(result, "markdown")
else None,
"word_count": word_count,
"links_count": links_count,
"images_count": images_count,
"execution_time": execution_time,
}
)
successful_count += 1
logger.info(
f"✅ Successfully crawled {url} in {execution_time:.2f}s"
)
else:
results.append(
{
"url": url,
"success": False,
"error_message": getattr(
result, "error_message", "Unknown error"
),
"execution_time": execution_time,
}
)
failed_count += 1
logger.warning(f"❌ Failed to crawl {url}")
except Exception as e:
error_msg = f"Error crawling {url}: {str(e)}"
logger.error(error_msg)
results.append(
{"url": url, "success": False, "error_message": error_msg}
)
failed_count += 1
# Format output
output_lines = [f"🕷️ Crawl4AI Results Summary:"]
output_lines.append(f"📊 Total URLs: {len(valid_urls)}")
output_lines.append(f"✅ Successful: {successful_count}")
output_lines.append(f"❌ Failed: {failed_count}")
output_lines.append("")
for i, result in enumerate(results, 1):
output_lines.append(f"{i}. {result['url']}")
if result["success"]:
output_lines.append(
f" ✅ Status: Success (HTTP {result.get('status_code', 'N/A')})"
)
if result.get("title"):
output_lines.append(f" 📄 Title: {result['title']}")
if result.get("markdown"):
# Show first 300 characters of markdown content
content_preview = result["markdown"]
if len(result["markdown"]) > 300:
content_preview += "..."
output_lines.append(f" 📝 Content: {content_preview}")
output_lines.append(
f" 📊 Stats: {result.get('word_count', 0)} words, {result.get('links_count', 0)} links, {result.get('images_count', 0)} images"
)
if result.get("execution_time"):
output_lines.append(
f" ⏱️ Time: {result['execution_time']:.2f}s"
)
else:
output_lines.append(f" ❌ Status: Failed")
if result.get("error_message"):
output_lines.append(f" 🚫 Error: {result['error_message']}")
output_lines.append("")
return ToolResult(output="\n".join(output_lines))
except ImportError:
error_msg = "Crawl4AI is not installed. Please install it with: pip install crawl4ai"
logger.error(error_msg)
return ToolResult(error=error_msg)
except Exception as e:
error_msg = f"Crawl4AI execution failed: {str(e)}"
logger.error(error_msg)
return ToolResult(error=error_msg)
def _is_valid_url(self, url: str) -> bool:
"""Validate if a URL is properly formatted."""
try:
result = urlparse(url)
return all([result.scheme, result.netloc]) and result.scheme in [
"http",
"https",
]
except Exception:
return False
|