sensei-eval

Version updated for https://github.com/CodeJonesW/sensei-eval to version v0.6.0.

• This action is used across all versions by 0 repositories.

Action Type

This is a Composite action.

Go to the GitHub Marketplace to find the latest changes.

Action Summary

The sensei-eval GitHub Action evaluates AI-generated educational content for quality using both deterministic checks and LLM-based scoring. It automates the detection of prompt quality regressions in CI workflows, enabling consistent content evaluation and quality assurance. Key capabilities include baseline generation, regression detection, and flexible configuration for content types, topics, and difficulty levels.

Release notes

Summary

• Updates @anthropic-ai/sdk from ^0.72.1 to ^0.78.0
• Fixes compatibility with Cloudflare Workers where the older SDK version fails with connection errors
• All 213 tests pass

Test plan

• npm run build succeeds
• npm test — all 213 tests pass
• Verify integration tests still pass in CI

🤖 Generated with Claude Code

• CodeJonesW
• GitHub Actions