evals-skills

Eval Skills

Skills that guide AI coding agents to help you build product-specific AI evals (not foundation model benchmarks).

These skills guard against common mistakes we've seen helping 50+ companies and teaching thousands of students in our AI Evals course.

Why skills for evals

There are many easily avoidable footguns in evals. These skills help you avoid them.

evals-start is the entry point. It looks at your situation and routes you to the right skill. Most of the time it will send you to one of these two:

• eval-audit, if you already have an eval pipeline. It inspects your setup and recommends next steps.
• error-discovery, if you have traces but haven't analyzed them yet. It builds an annotation interface and helps you sample traces. Watch Shreya's walkthrough.

Installation

Install with npx skills:

npx skills add https://github.com/ai-evals-course/evals-skills

Install one skill only:

npx skills add https://github.com/ai-evals-course/evals-skills --skill error-discovery

Check for updates:

npx skills check
npx skills update

Available skills

Skill	What it does
evals-start	Entry point. Routes to the skill that matches your situation
eval-audit	Audit an eval pipeline and surface problems with prioritized severity
error-discovery	Build a review app, select diverse samples, and organize your notes into failure modes
generate-synthetic-data	Create diverse synthetic test inputs using dimension-based tuple generation
write-code-eval	Write code checks for objective failure modes
write-judge-prompt	Design LLM-as-Judge evaluators for subjective quality criteria
validate-evaluator	Calibrate LLM judges against human labels using data splits, TPR/TNR, and bias correction
evaluate-rag	Evaluate retrieval and generation quality in RAG pipelines
build-review-interface	Build custom annotation interfaces for human trace review

The error-discovery skill

The error-discovery skill in the plugin is the most important. Error discovery involves qualitative and quantitative analysis of your traces to find failure modes. You should only write evals after doing this step.

This skill makes an AI agent run error analysis on a dataset. Point it at a JSONL/CSV/JSON file of LLM outputs or traces, and it:

1. Reads the dataset and figures out the content type (articles, agent traces, code, structured output, etc.).
2. Designs visual encoding based on what varies in the data. Uses Gestalt principles (color for categories, spacing for hierarchy, opacity for importance).
3. Builds a single-file HTML review app served by a Python stdlib server. No dependencies.
4. Clusters the data and picks a diverse initial sample (cluster reps + random picks).
5. Runs an interactive loop: monitors annotations, categorizes failure modes, proposes new samples to increase coverage.

You read and leave free-text notes. The agent sorts them into failure modes, tracks coverage, and picks new samples to fill gaps.

Once installed, point the agent at a dataset:

Can you help me do error analysis on traces.jsonl?

Write your own skills

These skills address common eval mistakes. Start here, then write skills for your own data and domain. Matt Pocock's writing-for-agents skill explains how to write instructions for agents.

Beyond these skills

These skills cover eval work that applies across projects. The AI Evals course also covers production monitoring, regression suites, and cost.