harness-selftest

Validate evaluation fixture integrity within CI/CD pipelines.

3|2|Updated Mar 17, 2026
One-click install
npx skills add https://github.com/broomva/skills --skill harness-selftest
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harness-selftest
Source: https://github.com/broomva/skills/tree/main/tests/skill_evals/fixtures/harness-selftest/skill
Command: npx skills add https://github.com/broomva/skills --skill harness-selftest

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill provides a controlled artifact for the evaluation harness to bind committed replay fixtures, ensuring that CI pipelines can verify prompt processing and grading logic without relying on external production tools.

Core Features & Use Cases

  • Evaluation Binding: Serves as a stable target for SHA-256 validation of test cases.
  • CI Gate Verification: Allows the evaluation harness to trigger and grade specific prompt sets to confirm that the testing infrastructure is functioning correctly.
  • Use Case: Developers use this to verify that the CI system correctly identifies and fails on modified prompt sets or invalid test fixtures.

Quick Start

Run the harness-selftest skill to validate the current evaluation fixture integrity.

Frequently Asked Questions about harness-selftest

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

FAQPage Schema
How do I validate automated evaluation fixtures in a CI pipeline?

A CI gate verification uses a stable target to trigger and grade specific prompt sets. This confirms the testing infrastructure correctly identifies and fails on modified prompt sets or invalid test fixtures.

What is a deterministic artifact for binding replay fixtures?

A deterministic artifact for binding replay fixtures is a controlled target used by an evaluation harness. It ensures CI pipelines can verify prompt processing and grading logic through SHA-256 validation of test cases.

Why does my CI pipeline fail when validating modified prompt sets?

CI pipelines fail when validating modified prompt sets to confirm the testing infrastructure is functioning correctly. This behavior verifies the evaluation harness properly triggers and grades prompt-response cycles against committed replay fixtures.

Can I use a non-functional target for regression analysis of prompt-response cycles?

Yes, you can use a non-functional target for regression analysis of prompt-response cycles. It acts as a stable artifact for the evaluation harness, ensuring CI pipelines can verify prompt processing and grading logic deterministically.

What are the limitations of using a test-only fixture for CI gate verification?

The limitation of a test-only fixture is that it requires the evaluation harness to be specifically configured to recognize it. It is a non-functional target solely for regression analysis, not for executing actual production tools.