Define eval criteria and track pass@k metrics for AI development.
npx skills add https://github.com/Luohaothu/everything-codex --skill eval-luohaothu
Please help me install this Agent Skill. Skill: eval Source: https://github.com/Luohaothu/everything-codex/tree/main/skills/eval Command: npx skills add https://github.com/Luohaothu/everything-codex --skill eval-luohaothu
Evals before coding establish expectations and enable continuous quality checks for AI systems.
Run an evaluation plan by defining evals and executing checks with the /eval commands.
High-intent search queries and answers about installing and using this skill.
To define eval criteria before coding, you establish formal evaluation expectations and execute checks using standardized commands. This eval-driven approach enables continuous quality assurance across feature development and model updates.
Eval-driven development is a process where you define capability and regression evaluations before writing code. It ensures AI quality by establishing expectations early and tracking pass@k metrics to monitor for regressions across release cycles.
Yes, the evaluation framework supports regression testing for model updates. It tracks pass@k metrics and generates reports to monitor regressions, ensuring reliable capabilities are shipped across release cycles.
Yes, you can use multiple grader types for AI capability evaluations. The framework supports various grader types and standardized workflows to assess and ensure the quality of AI-driven development.
The best way to store and manage evaluation data is using a formal eval framework with standardized workflows. It provides eval storage and tracks pass@k metrics to maintain continuous quality checks for AI systems.
rongaredeEvaluate Claude Code sessions with capability and regression evals.
vinayakgImplement eval-driven development with code-based, model-based, and human grading.
steve-pieceOrchestrate CI/CD pipelines, E2E testing, and design-system compliance gates.
nirukk52Monitor Encore.ts and SvelteKit logs to diagnose drift and verify run events.
