# evals-skills: a set by ai-evals-course

The ai-evals-course/evals-skills repository on GitHub: 9 cards with a description and an install command. They work in Claude Code, Codex, Cursor — 8 platforms in total.

9 cards

- HTML: https://affpapa.org/en/nevibecode/by/ai-evals-course/evals-skills
- Instructions for agents: https://affpapa.org/en/nevibecode/agents.md

- [write-judge-prompt](https://affpapa.org/en/nevibecode/skills/write-judge-prompt.md) — skill · ai-evals-course · Claude Code, Codex, Cursor +5 · 1300 installs — write-judge-prompt walks you through one binary Pass/Fail LLM-as-Judge evaluator for a subjective failure mode: tone, faithfulness, relevance.
- [validate-evaluator](https://affpapa.org/en/nevibecode/skills/validate-evaluator.md) — skill · ai-evals-course · Claude Code, Codex, Cursor +5 · 1300 installs — The validate-evaluator skill checks whether an LLM judge agrees with expert labels and measures TPR/TNR on a held-out set.
- [generate-synthetic-data](https://affpapa.org/en/nevibecode/skills/generate-synthetic-data.md) — skill · ai-evals-course · Claude Code, Codex, Cursor +5 · 1200 installs — generate-synthetic-data — a skill that builds varied synthetic test queries for LLM pipeline evaluation when real data is sparse.
- [eval-audit](https://affpapa.org/en/nevibecode/skills/eval-audit.md) — skill · ai-evals-course · Claude Code, Codex, Cursor +5 · 1200 installs — eval-audit audits LLM evaluation pipelines to uncover unvalidated judges, vanity metrics, and flawed error analysis.
- [error-discovery](https://affpapa.org/en/nevibecode/skills/error-discovery.md) — skill · ai-evals-course · Claude Code, Codex, Cursor +5 · 1200 installs — error-discovery — a skill for interactive error analysis on a dataset of LLM outputs: it builds a review UI, picks diverse samples and helps organize failure modes.
- [build-review-interface](https://affpapa.org/en/nevibecode/skills/build-review-interface.md) — skill · ai-evals-course · Claude Code, Codex, Cursor +5 · 1100 installs — build-review-interface builds a custom browser-based annotation interface to review LLM traces and record human feedback.
- [evaluate-rag](https://affpapa.org/en/nevibecode/skills/evaluate-rag.md) — skill · ai-evals-course · Claude Code, Codex, Cursor +5 · 1100 installs — evaluate-rag guides separate evaluation of RAG retrieval and generation, metric selection, and document chunking tests.
- [evals-start](https://affpapa.org/en/nevibecode/skills/evals-start.md) — skill · ai-evals-course · Claude Code, Codex, Cursor +5 · 959 installs — evals-start routes an evaluation request to the right specialised skill.
- [write-code-eval](https://affpapa.org/en/nevibecode/skills/write-code-eval.md) — skill · ai-evals-course · Claude Code, Codex, Cursor +5 · 535 installs — write-code-eval is a Claude Code skill for building code evaluators with objective rules.

## Narrow this list
Parameters: `?platform=<claude-code|codex|cursor|copilot|gemini-cli|windsurf|…>`, `&topic=<…>`, `&sort=installs|stars|new|updated` — values in https://affpapa.org/en/nevibecode/agents.md

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
