Integrations
Crawl4AI
The open-source Crawl4AI renderer, pinned and isolated in its own environment.
Crawl4AI 0.9.4 runs in a separate Python environment with its own browsers, so its dependencies never touch Frankensurf’s.
- Pinned to an exact package lock; every dependency is hash-checked before each run.
- Uses its own browser cache, never the shared Playwright one.
- Any change to the install makes the provider unavailable instead of running unchecked code.
Set up
Section titled “Set up”bash scripts/install-crawl4ai-provider.shThe installer creates ~/.local/share/frankensurf/crawl4ai-venv and ends with a live render check. Move it with FRANKENSURF_CRAWL4AI_DIR, FRANKENSURF_CRAWL4AI_PYTHON and FRANKENSURF_CRAWL4AI_BROWSERS_PATH.
Use it
Section titled “Use it”It joins the default route by itself once set up. To use it on its own:
page = await web.read( "https://example.com/products/2231", provider="crawl4ai",)frankensurf read https://example.com/products/2231 --provider crawl4ai{ "tool": "read", "arguments": { "url": "https://example.com/products/2231", "provider": "crawl4ai" } }Details
Section titled “Details”| Property | Value |
|---|---|
| Provider ID | crawl4ai |
| Rung | T2 |
| Cost | Free |
| Tested | Live health check |
| Docs | docs.crawl4ai.com |