
from everything-evenhub43
A reproducible test harness skill that runs implementer and verifier subagents to validate other everything-evenhub skills end-to-end using structured checklist
The harness skill orchestrates reproducible tests for everything-evenhub skill packages. It automates a two-step validation: dispatching an implementer subagent to execute a simulated user request and then a verifier subagent to independently check outputs against a checklist. The focus is on determinism and auditability for skill quality assurance.
Invoke this skill when you need to validate that a skill produces correct outputs, when running CI for skill changes, or when auditing implementation quality. Typical scenarios: running quickstart/build-and-deploy tests, UI generation checks, or cross-check audits.
This orchestration skill expects agent runtimes capable of spawning subagents and running shell commands (Claude Code, sonnet-capable subagents, and specialized verifier agents).
This skill has not been reviewed by our automated audit pipeline yet.