Releases: alibaba/page-agent
Releases · alibaba/page-agent
🐙 Chrome extension v0.1.15
fix: add button to clear saved configuration from the error boundary
🌟 v1.5.2
Breaking Changes
data-browser-use-ignore→data-page-agent-ignore- DOM ignore attribute renamed to match the project identity- Config types restructured -
PageAgentConfigsplit intoAgentConfig+PageAgentCoreConfig; config definitions moved fromconfig/index.tstotypes.ts - Zod v3/v4 dual support - Libraries now accept both
zod@^3.25andzod@^4.0as peer dependencies
Features
- Experimental
llms.txtsupport - Agent can fetch and include a site'sllms.txtin context. Enable viaexperimentalLlmsTxt: true
Improvements
- Default
maxStepschanged from 20 to 40 for better for complex tasks out of the box - Added 400ms wait between agent steps for page reactions
- Increased click wait time (100ms → 200ms) for more reliable interactions
- Removed debug
console.logstatements from scroll actions - Reset observations on new task start
- Improved logging across packages
🐙 Chrome extension v0.1.12
fix: crash when webgl2 not available
🐙 Chrome extension v0.1.11
PageAgent 1.5.1
- Advanced config panel - New collapsible section exposing Max Steps, System Instruction, and experimental
llms.txttoggle - Streamlined User Auth Token description
- Moved testing API notice below auth token section
🌟 v1.4.0
Features
- Update Terms of Use and Privacy Policy
- Robust tool-call validation - Action inputs are now validated against tool schemas individually, producing clear error messages (e.g.
Invalid input for action "click_element_by_index") instead of unreadable union parse errors - Primitive action input coercion - Small models that output
{"click_element_by_index": 2}instead of{"click_element_by_index": {"index": 2}}are now auto-corrected using tool schemas - Qwen model updates - Added
qwen3.5-plusas the default free testing model; disabledenable_thinkingfor Qwen models to avoid incompatible responses - Updated default LLM endpoint - Migrated demo and extension to a new testing endpoint with legacy endpoint auto-migration
Improvements
- Unified zod imports (
* as z) across all packages for consistency - Better Zod error formatting with
z.prettifyError()in LLM client - Exported
InvokeErrorandInvokeErrorTypeas values (not just types) from@page-agent/llms - Exported
SupportedLanguagetype from@page-agent/core
🐙 Chrome extension v0.1.8
- Update PageAgent to 1.4.0
- Language setting - Added language selector (System / English / 中文) in config panel
- UI makeover - New empty state with breathing glow and typing animation; ai-motion glow overlay while running; refined focus styles
- Testing endpoint notice - Shows terms of use notice when using the free testing API
- Legacy endpoint migration - Auto-migrates old Supabase testing endpoint to new endpoint on startup
🧪 Chrome extension v0.1.7
- Update page agent version
- Update UX
- Add locales
🌟 v1.3.0
Breaking Changes
- Lifecycle:
stop()vsdispose()- Newstop()method to cancel the current task while keeping the agent reusable.dispose()is now terminal — a disposed agent cannot be reused. This affects bothPageAgentCoreandPanelAgentAdapter.
Features
- Panel action button - The panel button now morphs between Stop (■) and Close (X) based on agent status
- Error history - Errors and max-step failures are now recorded in
historyasAgentErrorEvent, making post-task analysis more complete
Bug Fixes
- AbortError handling -
AbortErroris no longer retried by the LLM client, and shows a clean "Task stopped" message instead of a raw error stack
🌟 v1.2.0
What's Changed
- Improve prompts
- Clean up code
- Add a
observephase in step - Major update of extension
Full Changelog: v1.1.0...v1.2.0
🧪 Chrome extension v0.1.5
- Fixed a lot of cross-thread problems