fix: correct Lightpanda fallback docs — remove nonexistent PDF/upload/clipboard actions
Hermes has no browser PDF, file upload, or clipboard tools. The fallback mechanism only covers commands in _FALLBACK_ELIGIBLE (open, snapshot, screenshot, eval, click, fill, scroll, back, press, console, errors). The original docs described Lightpanda's general limitations, not Hermes's actual behavior.
This commit is contained in:
parent
0a81935a9d
commit
a871948d8d
|
|
@ -353,7 +353,7 @@ AGENT_BROWSER_ENGINE=lightpanda
|
|||
|
||||
Hermes drives Lightpanda through `agent-browser` over CDP, the same way it drives local Chrome.
|
||||
|
||||
**Automatic Chrome fallback.** Lightpanda doesn't yet cover everything Chrome does, so the integration is non-disruptive: Lightpanda handles the actions it supports, and Hermes transparently retries on Chrome for anything it doesn't. The supported set covers the core agent workflow — navigate, snapshot, click, type, scroll, back, press, and eval. Actions that fall back to Chrome include screenshots (Lightpanda has no graphical renderer), PDF generation, file uploads, and clipboard operations. Because Lightpanda skips visual rendering, `browser_vision` is pre-routed straight to Chrome.
|
||||
**Automatic Chrome fallback.** Lightpanda doesn't yet cover everything Chrome does, so the integration is non-disruptive: Lightpanda handles the actions it supports, and Hermes transparently retries on Chrome for anything it doesn't. The supported set covers the core agent workflow — navigate, snapshot, click, type, scroll, back, press, and eval. Screenshots also fall back to Chrome because Lightpanda has no graphical renderer; `browser_vision` is pre-routed straight to Chrome for the same reason.
|
||||
|
||||
### Local Chromium-family browser via CDP (`/browser connect`)
|
||||
|
||||
|
|
|
|||
Loading…
Reference in New Issue