Skill Tester

Helps create rubrics and test scenarios for evaluating AI skills, and evaluates conversation traces against those rubrics. Triggers when the user wants to test a skill, write a rubric for a skill, evaluate whether a skill is working well, define quality criteria for a skill, or assess a conversation where a skill was used.

48 stars 8 forksSoftware Development

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add harvard-lil/lawskills-hub --skill skill-tester
Manual install
git clone --depth 1 https://github.com/harvard-lil/lawskills-hub.git /tmp/lawskills-hub
mkdir -p ~/.claude/skills
cp -r /tmp/lawskills-hub/skills/skill-developer/skill-tester ~/.claude/skills/skill-tester

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Skill Tester

You help a **skill developer** define quality criteria for a skill, create test scenarios, and evaluate whether the skill performs well in practice. You work with a rubric format that supports both automated and human evaluation.

## Tone

Methodical and precise. You are a quality engineer for pedagogical experiences. Be concrete about what "good" looks like and honest about what falls short.

## Background: The Rubric Format

Each skill can have a rubric at `_hub/rubric.yaml`, in a `_hub/` directory beside its `SKILL.md`. Files in `_hub/` are read by the hub and test harness and are not packaged with the skill. A rubric defines:

- **Criteria**: What "good" looks like for this skill, split into:
  - **Structural criteria**: Concrete, checkable behaviors (e.g., "agent asks about context before giving advice"). These can be evaluated deterministically.
  - **Pedagogical criteria**: Subjective quality judgments (e.g., "agent coaches rather than tells"). These require human or LLM evaluation.
- **Anti-patterns**: Things the agent must never do. Violations are automatic failures.
- **Test scenarios**: Scripted user interactions with expected agent behaviors.

Here is the rubric schema:
Read the full skill on GitHub

Details

Skill name
skill-tester
Path
skills/skill-developer/skill-tester/SKILL.md
Repo updated
16 October 2026

FAQ

What does the Skill Tester skill do?

Helps create rubrics and test scenarios for evaluating AI skills, and evaluates conversation traces against those rubrics. Triggers when the user wants to test a skill, write a rubric for a skill, evaluate whether a skill is working well, define quality criteria for a skill, or assess a conversation where a skill was used.

How do I install the Skill Tester skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add harvard-lil/lawskills-hub --skill skill-tester. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Skill Tester popular?

It lives in harvard-lil/lawskills-hub, which has 48 GitHub stars and 8 forks.

Related skills

Cle Meta

48

harvard-lil/lawskills-hub

Always-on assistant for practicing attorneys focused on professional development and continuing education. Triggers on professional development, skill building, learning new practice areas, client communication improvement, CLE planning, career development, or practice management improvement.

Software Development

Client Email Coach

48

harvard-lil/lawskills-hub

Triggers when a practicing attorney wants feedback on a draft email to a client, needs help improving client communication, or wants coaching on tone, clarity, or risk management in written client communications.

Productivity & Notes

Development Plan

48

harvard-lil/lawskills-hub

Triggers when a practicing attorney wants to create a professional development plan, set career goals, identify skill gaps, plan their CLE activities for the year, or think strategically about their career trajectory.

Software Development

D365fo CLI

49

dynamics365ninja/d365fo-cli

D365 Finance & Operations X++ AI development skill powered by the d365fo CLI. Use whenever the user is working in a D365 F&O X++ project: writing, reading, searching, reviewing, or debugging classes, tables, forms, CoC extensions, event handlers, entities, security, batch jobs, business events, labels, or any AOT artifact — including bug investigation and code-reading tasks, not just new authoring. Prefer `d365fo search`/`get`/`find` over Grep/Glob when locating or inspecting AOT objects. Points to the topic-specific d365fo skills installed next to it.

Software Development26

Skill Reviewer

48

harvard-lil/lawskills-hub

Reviews and evaluates an existing SKILL.md for quality, pedagogical soundness, persona compliance, and agent-readiness. Triggers when the user wants feedback on a skill they've written, wants to improve an existing skill, or wants to check whether a skill meets the Legal Ed Skills Hub standards.

Software Development

Debug Playbook

49

toffyui/ccteams

Operational playbook for systematic debugging — hypothesis ledgers, discriminating experiments, mechanism sentences, minimal fixes, sibling hunts. Read in full at the start of every bug investigation.

Software Development

Browse the most popular Claude skills, all software development skills, or write your own with the skills generator.