skill-harness

Name: skill-harness
Availability: InStock
Author: ginmp

Audit, validate, and harden ChatGPT or agent skills across multiple modes.

Updated Apr 27, 2026

Search Tags:#validation #audit #scenarios #gates #package #harness

Ready to run

npx skills add https://github.com/ginmp8/rhapsodia --skill skill-harness

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: skill-harness
Source: https://github.com/ginmp8/rhapsodia/tree/main/skills/skill-harness
Command: npx skills add https://github.com/ginmp8/rhapsodia --skill skill-harness

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

Build a repeatable harness to audit, validate, harden, compare, and package existing ChatGPT or agent skills across multiple modes.

Core Features & Use Cases

Supports auto, context, and full harness modes for evidence-based improvement
Enables audit, plan, apply, validation, and packaging workflows for target SKILL.md
Reads references, scripts, assets, and evals to gather evidence and build durable reports

Quick Start

Run the harness on a target skill folder to inventory, audit, plan, and package a validated skill.zip

Frequently Asked Questions about skill-harness

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

FAQPage Schema

How do I audit and validate existing ChatGPT skills?▼

To audit and validate ChatGPT skills, you can run a repeatable harness that reads SKILL.md files, scripts, and assets to gather evidence and generate durable validation reports.

What is the best way to harden an agent skill package?▼

The best way to harden an agent skill package is to apply a deterministic harness that enforces entry points, gates, and metrics, ensuring bounded mutations across the target skill folder.

Can I package a validated agent skill into a zip file automatically?▼

Yes, you can package a validated agent skill automatically by running the harness in package mode, which inventories, audits, and validates the target folder before exporting a durable skill.zip.

What modes are supported for running a skill validation harness?▼

Skill validation harnesses support auto, context, and full modes, enabling you to execute audit, plan, apply, validation, and packaging workflows for your target SKILL.md.

Does the skill harness require external dependencies to evaluate scripts and references?▼

No, the skill harness operates without external dependencies, reading references, scripts, assets, and evals directly to gather evidence and build reports for the target skill package.