What's new?

• fix: do not add useless comments/verdicts [ENG-1596]
• fix: post code review without line comments if there are hunk line errors [ENG-1597]

Full Changelog:v2.2.0...v2.2.1

What's Changed

• fix comment issues + add support to o1-mini by@lfsevergnini in#18

Full Changelog:v2.1.2...v2.2.0

Contributors

What's Changed

• Add temperature + Tweak prompt for better verdicts by@lfsevergnini in#17

Full Changelog:v2.1.1...v2.1.2

Contributors

What's Changed

• fix: prevent errors when LLM returns comments on lines not in the diff by@lfsevergnini in#16

Full Changelog:v2.1.0...v2.1.1

Contributors

What's Changed

• Add support to optional parameters (context, auto approve, etc) by@lfsevergnini in#15

Full Changelog:v2.0.2...v2.1.0

Contributors

Full Changelog:v2.0.1...v2.0.2

Full Changelog:v2.0.0...v2.0.1

What's Changed

• breaking change: AI Code Reviewer V2 by@lfsevergnini in#14

Full Changelog:v1.0.0...v2.0.0

Contributors

Improvements

• Upgraded to Node 20
• Changed default GPT model
• Implemented failure handling for non-200 responses from OpenAI
• Added comprehensive logging throughout the application
• Enabled JSON mode for OpenAI responses
• Upgraded OpenAI package to latest version

New Features

• Added ability to approve reviews based on environment variable
• Implemented check for existing reviews before deciding on incremental review

Bug Fixes

• Strip markdown formatting from responses
• Replace console.log with core.info for better GitHub Actions integration

Other Changes

• Various code refactoring and cleanup

Full Changelog:v0.0.5...v1.0.0