fetch_url.py

from urllib.parse import urlsplit


TOOL = {
    'id': 'fetch_url',
    'name': '网页信息提取',
    'description': '读取公开网页的标题、描述、正文摘要和主要链接信息；需要了解指定网页内容时使用。',
    'parameters': {
        'type': 'object',
        'properties': {
            'url': {'type': 'string', 'description': '公开 HTTP 或 HTTPS 网页地址', 'maxLength': 2000},
        },
        'required': ['url'],
        'additionalProperties': False,
    },
}


async def run(arguments, context):
    url = str(arguments.get('url') or '').strip()[:2000]
    try:
        parsed = urlsplit(url)
    except ValueError:
        parsed = None
    if parsed is None or parsed.scheme not in {'http', 'https'} or not parsed.hostname:
        return {'ok': False, 'error': '请输入有效的 HTTP 或 HTTPS 网页地址'}
    data = await context['http_json'](
        'https://api.microlink.io/',
        {'url': url, 'audio': 'false', 'video': 'false', 'screenshot': 'false'},
    )
    if not isinstance(data, dict) or data.get('status') != 'success':
        message = data.get('message') if isinstance(data, dict) else None
        return {'ok': False, 'error': str(message or '网页信息提取失败')[:300]}
    page = data.get('data') if isinstance(data.get('data'), dict) else {}
    image = page.get('image') if isinstance(page.get('image'), dict) else {}
    logo = page.get('logo') if isinstance(page.get('logo'), dict) else {}
    return {
        'ok': True,
        'url': page.get('url') or url,
        'title': page.get('title') or '',
        'description': page.get('description') or '',
        'publisher': page.get('publisher') or '',
        'author': page.get('author') or '',
        'language': page.get('lang') or '',
        'date': page.get('date') or '',
        'image': image.get('url') or '',
        'logo': logo.get('url') or '',
    }