Skip to content
mcp/skillhub

Agent Skill

Agent Evals skill

agent-evalssickn33/agentic-awesome-skillsplugins/agentic-awesome-skills-claude/skills/agent-evals/SKILL.md

Build automated evaluation suites for AI agents using golden datasets, rubrics, and regression gates. Use when shipping agent features, validating prompt changes, or gating deployments on quality.

48kMIT

always on

38 tokname + description

once used

3,084 tokSKILL.md body

bundled files

0

repo stars

48kGitHub

Install the Agent Evals skill

Skills CLI (Claude Code, Codex, Cursor and others)

npx skills add https://github.com/sickn33/agentic-awesome-skills --skill agent-evals

Claude Code, by hand (all projects)

git clone --depth 1 https://github.com/sickn33/agentic-awesome-skills.git /tmp/agentic-awesome-skills
mkdir -p ~/.claude/skills
cp -R /tmp/agentic-awesome-skills/plugins/agentic-awesome-skills-claude/skills/agent-evals ~/.claude/skills/agent-evals

For one project only, copy the folder into .claude/skills/ in that project instead. The agent picks it up when a task matches its description.

What it costs your context

Agent Skills load in three steps. Tokens counted with the o200k tokenizer; your model's count will differ a little.

WhenWhat loadsCost
Every sessionName and description38 tok
When the agent uses itThe SKILL.md instructions3,084 tok
Only if openedNo bundled files—

Until it is used, agent-evals costs about 36× less than the median MCP server we measured (1,362 tokens on every turn). Installing all 150 skills from sickn33/agentic-awesome-skills would load 5,255 tokens of descriptions. Skills vs MCP study

Checks

Our checks found nothing to flag: it follows the Agent Skills specification's naming and size guidelines and doesn't pipe downloads into a shell.

Checked 2026-10-22 · method

What's inside

The headings of its SKILL.md. Read the full file on GitHub.

  • Agent Evals
  • When to Use This Skill
  • Prerequisites
  • Evaluation contract
  • Evaluation Layers
  • Unit Evals — Prompt-Level Correctness
  • Tool Evals — Decision Quality
  • End-to-End Evals — Multi-Step Tasks
  • Safety Evals — Adversarial Resistance
  • LLM-as-Judge Scoring
  • Promptfoo Configuration
  • CI/CD Integration
  • GitHub Actions
  • Makefile Targets
  • Tracking Eval Drift
  • Best Practices
  • Related Skills
  • Limitations
  • Example

More skills from sickn33/agentic-awesome-skills

All skills in this repo
  1. 00-andruia-consultantsickn33/agentic-awesome-skills

    Usar al iniciar o mejorar un proyecto de IA: diagnostica código y restricciones, propone skills existentes y redacta tareas y un plan de implementación en español.

    48k40 tok idle · 1,134 in use
  2. 007sickn33/agentic-awesome-skills

    Security audit, hardening, threat modeling (STRIDE/PASTA), Red/Blue Team, OWASP checks, code review, incident response, and infrastructure security for any project.

    48k839 tok idle · 259 in use
  3. 10-andruia-skill-smithsickn33/agentic-awesome-skills

    Usar para crear o mejorar una skill de Andru.ia: define entradas y resultados, redacta SKILL.md con ejemplos y límites, valida la estructura y prepara una PR.

    48k46 tok idle · 907 in use
  4. 20-andruia-niche-intelligencesickn33/agentic-awesome-skills

    Usar tras definir el sector de un proyecto: prepara un dossier en español con contexto, normas aplicables, patrones UX y riesgos, distinguiendo fuentes verificadas de hipótesis.

    48k45 tok idle · 1,264 in use
  5. 2slides-ppt-generatorsickn33/agentic-awesome-skills

    AI-powered presentation generation via the 2slides API — create slides from text, match a reference image style, summarize documents into decks, add AI voice narration, and export pages/audio. Use for any "make slides", "create a deck", or "slides from this document" request.

    48k764 tok idle · 594 in use
  6. 360-feedback-systemsickn33/agentic-awesome-skills

    360 feedback register: reviewer, subject, review cycle, visibility, due date and score, as CSV, SQL, JSON Schema or Notion on request. Use for 360 reviews or peer feedback cycles.

    48k46 tok idle · 4,900 in use
  7. 3d-web-experiencesickn33/agentic-awesome-skills

    Use when building or debugging web 3D scenes, product configurators or portfolios with Three.js, React Three Fiber or Spline; review assets, loading, controls, fallbacks and performance.

    48k45 tok idle · 2,579 in use
  8. aas-compose-stacksickn33/agentic-awesome-skills

    Preserve the coding agent's explicit AAS skill selection as a validated stack and optional evidence.

    48k22 tok idle · 468 in use