This service implements the protocol expected by Open WebUI ExternalWebLoader:
-
Request:
- POST {EXTERNAL_WEB_LOADER_URL}
- Header: Authorization: Bearer {EXTERNAL_WEB_LOADER_API_KEY}
- Body: {"urls": ["https://...", ...]}
-
Response:
- [{"page_content": "...", "metadata": {"source": "..."}}, ...]
Environment:
- WEB_LOADER_API_KEY: api key (optional)
- PLAYWRIGHT_WS_URL: ws://playwright:3000
- NAV_TIMEOUT_MS: default 12000
- WAIT_UNTIL: domcontentloaded (recommended)
- BLOCK_RESOURCE_TYPES: image,font,media (recommended)
- MAX_CONCURRENCY: 3
本服务实现了 Open WebUI ExternalWebLoader 所期望的协议:
-
请求:
- POST {EXTERNAL_WEB_LOADER_URL}
- Header: Authorization: Bearer {EXTERNAL_WEB_LOADER_API_KEY}
- Body: {"urls": ["https://...", ...]}
-
响应:
- [{"page_content": "...", "metadata": {"source": "..."}}, ...]
环境变量:
- WEB_LOADER_API_KEY:API key(可选)
- PLAYWRIGHT_WS_URL:ws://playwright:3000
- NAV_TIMEOUT_MS:默认 12000
- WAIT_UNTIL:domcontentloaded(推荐)
- BLOCK_RESOURCE_TYPES:image,font,media(推荐)
- MAX_CONCURRENCY:3