eval-framework
CommunityEvaluate AI agent reliability with pass@k metrics
AuthorCheggin
Version1.0.0
Installs0
System Documentation
What problem does it solve?
Evaluates AI agent reliability by focusing testing on changes that affect behavior, reducing wasted compute and surfacing flaky or brittle interactions early.
Core Features & Use Cases
- Diff-driven eval execution: runs only evals impacted by a git diff to minimize CI time.
- Pass@k and pass^k analytics: measures partial and full success rates with per-eval diagnostics.
- CI integration: hooks and dashboards to detect regressions and track performance across releases.
- Real-world use cases include validating agent safety, stability after feature changes, and regression testing of capabilities.
Quick Start
Run the evaluation suite against your agent changes to measure pass@k metrics.
Dependency Matrix
Required Modules
None requiredComponents
Standard package💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: eval-framework Download link: https://github.com/Cheggin/request-for-startups/archive/main.zip#eval-framework Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.