Skip to content

Upgrade browser_use from v0.7.10 to Latest Version and Implement Breaking Changes #12

@monkscode

Description

@monkscode

Issue Type:
Task / Technical Debt

Priority:
Medium

Description:
Objective: Update the browser_use library from version 0.7.10 to the latest stable release and implement necessary code changes to maintain compatibility.

Current State:

Current version: browser_use==0.7.10 (defined in src/backend/requirements.txt)
Used in: Browser automation service for vision AI element identification
Key integration points:
tools/browser_use_service.py - Main service entry point
tools/browser_use_tool.py - CrewAI tool interface
browser-service/browser_service/tasks/workflow.py - Agent implementation
Multiple API endpoints and metrics tracking
Scope of Work:

Version Analysis

Review browser_use changelog/release notes from v0.7.10 to latest
Identify breaking changes, deprecated APIs, and new features
Document migration requirements
Code Impact Assessment

Audit all browser_use imports and API usage across codebase
Check Agent class initialization and method signatures
Verify custom actions compatibility (ENABLE_CUSTOM_ACTIONS)
Review locator extraction and validation logic
Implementation

Update src/backend/requirements.txt with new version
Refactor affected code in:
browser_service/tasks/workflow.py (Agent usage)
browser_service/agent/ (custom actions)
tools/browser_use_tool.py (API client)
Update configuration if new options available
Ensure Windows UTF-8 compatibility maintained
Testing & Validation

Test workflow endpoint (POST /workflow)
Verify batch processing functionality
Validate locator generation (ID, CSS, XPath, Playwright selectors)
Test with both Browser Library and SeleniumLibrary modes
Confirm metrics tracking still works
Documentation:

Update CHANGELOG.md with upgrade details
Document any new configuration options
Update inline comments if API usage changed
Affected Components:

Browser automation service (Flask API)
Task processor and workflow execution
Agent management and custom actions
Locator generation and validation
Metrics tracking system
Dependencies:

Requires playwright compatibility check
May affect robotframework-browser[bb] integration
Verify langchain-google-genai compatibility
Acceptance Criteria:

[ ] browser_use upgraded to latest stable version
[ ] All existing functionality works without regression
[ ] Tests pass for both Browser and Selenium library modes
[ ] No breaking changes in API endpoints
[ ] Documentation updated
[ ] CHANGELOG.md reflects the upgrade
Risk Assessment:

Medium risk: Core dependency for vision-based automation
Mitigation: Test in isolated environment first, maintain rollback capability

Metadata

Metadata

Assignees

No one assigned

    Labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions