test-gate

Enforce verification gate decisions by comparing planned and executed test commands.

Updated Apr 9, 2026
One-click install
npx skills add https://github.com/ldmrepo/agent-harness --skill test-gate
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: test-gate
Source: https://github.com/ldmrepo/agent-harness/tree/main/template/.claude/skills/test-gate
Command: npx skills add https://github.com/ldmrepo/agent-harness --skill test-gate

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill standardizes and enforces repository verification gates to prevent tasks and sessions from being marked as verified or approved without appropriate executed test evidence, ensuring consistent, auditable gate decisions across sessions.

Core Features & Use Cases

  • Gate Depth Determination: Decide whether smoke, task-level, integration, full-verify, or extended checks are required based on task type and repository policy.
  • Evidence Validation: Compare planned versus executed test commands, assess execution results and report quality, and detect missing or contradictory evidence.
  • Artifact Updates & Approval Impact: Record explicit gate outcomes into session summaries and current task artifacts and surface approval guidance (passed, failed, pending, not_evaluable) for reviewers.
  • Use Case: Apply when ending a coding session, evaluating a feature or bugfix for verified status, or enforcing core/shared code verification before approval.

Quick Start

Use the test-gate skill to evaluate executed verification commands, classify the gate outcome, and update the session summary and current task record accordingly.

Frequently Asked Questions about test-gate

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

FAQPage Schema
How do I enforce a verification gate to prevent unverified code session approvals?

To enforce a verification gate, you compare planned versus executed test commands, validate execution evidence quality, and write explicit gate outcomes into session summaries to drive approval decisions.

What is a quality gate check for coding session close and pre-merge reviews?

A quality gate check for session close evaluates whether required tests ran and whether evidence supports approval, classifying outcomes as passed, failed, pending, or not_evaluable for reviewers.

How do I validate test execution evidence before approving a bugfix or feature completion?

You validate test execution evidence by comparing planned versus executed test commands, assessing execution results and report quality, and detecting missing or contradictory evidence before approval.

Can I determine the required test depth for core or shared code changes before merge?

Yes, you can determine gate depth by deciding whether smoke, task-level, integration, full-verify, or extended checks are required based on task type and repository policy for core code changes.

Why does my session approval fail when test evidence is missing or contradictory?

Session approvals fail because the verification gate detects missing or contradictory evidence, classifying the gate outcome as failed or not_evaluable and preventing unverified tasks from being approved.

Does test-gate work without external dependencies for repository verification?

Yes, test-gate operates with no external dependencies, applying its internal logic to repository verification, approval, and session-close scenarios across various task types.