skills.rest
servitola
servitolaCOMMUNITY
Explore all skills by this author →
GitHub

skill-testing

Design evaluation scenarios, run parallel subagent tests, and grade Claude Code skills.

12|2|Updated Feb 2, 2020
Search Tags:#automation#orchestration#claude-code#grading#skill-testing#scenario-design
One-click install
npx skills add https://github.com/servitola/dotfiles --skill skill-testing-servitola
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-testing
Source: https://github.com/servitola/dotfiles/tree/main/claude-code/skills/skill-testing
Command: npx skills add https://github.com/servitola/dotfiles --skill skill-testing-servitola

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Automates end-to-end evaluation of Claude Code skills by designing evaluation scenarios, orchestrating parallel subagent runners, and grading results with a consolidated report.

Core Features & Use Cases

  • Scenario design: compose repeatable skill tests from templates and references.
  • Orchestration: run multiple runners against a no-skill baseline to measure added value.
  • Reporting: produce evidence-rich verdicts and shareable reports for compliance and QA.

Quick Start

Design and run end-to-end skill tests for a given skill using the provided scenario templates.

Frequently Asked Questions about skill-testing

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate end-to-end evaluation of Claude Code skills?▼

To automate end-to-end evaluation of Claude Code skills, you can use a structured workflow that designs evaluation scenarios, orchestrates parallel subagent runners, and grades results with evidence-based reports.

What is the best way to design repeatable test scenarios for Claude Code?▼

The best way to design repeatable test scenarios for Claude Code is to compose them from predefined templates and references, ensuring reproducible results through phase-based mapping and journal-based traces.

Can I measure the added value of my skill against a baseline without it?▼

Yes, you can measure the added value of your skill by orchestrating multiple parallel runners against a no-skill baseline, which isolates the performance impact of the skill being tested.

How do I generate evidence-based reports for skill grading?▼

You generate evidence-based reports for skill grading by applying a standardized grading and reporting format across quick, design, run, and full execution modes to produce consolidated compliance verdicts.

Does skill testing require any external dependencies to run?▼

No, skill testing requires no external dependencies to run, as it operates self-contained by reading SKILL.md files and applying its own structured workflow for scenario design and orchestration.

Related Skills

Explore More Software Engineering Skills→
MiniMax-AIMiniMax-AI
2.9k

skill-creator

Generate Skill Units with YAML frontmatter, Markdown body, and bundled resources.

Official
Intermediate
MihaiOnSoftwareMihaiOnSoftware
1

write-validation-report

Aggregate validation results into a dated pass/fail Markdown report.

Community
Basic
hotriluanhotriluan
3

mcp-management

Manage MCP servers and execute tools, prompts, and resources across configured hosts.

Community
Intermediate
AshesOfTheUndeadAshesOfTheUndead

skill-creator

Develop, evaluate, and optimize AI skills with iterative testing and Python-based metrics.

Community
Advanced