How does this improve team collaboration?

It builds trust between the human developer and the AI by ensuring every commit or PR is backed by verified results rather than assumptions.

Why is this skill necessary for Claude Code?

It prevents 'hallucinated' success reports by forcing the AI to provide fresh command-line evidence before claiming a task is done, maintaining high trust.

Can I use this for build checks as well as unit tests?

Yes, it covers build success, linter cleanliness, regression tests, and requirement checklists to ensure comprehensive project health.

What happens if a verification command fails?

The skill prevents the AI from making a success claim and requires it to report the actual status with evidence of the failure for further debugging.

Completion Verification Guardrail

Name: Completion Verification Guardrail
Author: julianromli

byjulianromli

•

101

セキュリティとテスト

Enforces rigorous verification protocols to ensure work is empirically validated before making status claims or creating pull requests.

概要

The Completion Verification Guardrail implements an 'evidence-before-claims' policy that prevents premature task completion reports within Claude Code. By establishing a mandatory gate function, this skill requires the identification and execution of fresh verification commands—such as test suites, build scripts, or linters—before any work can be declared finished. It effectively eliminates 'hallucinated' success by replacing assumptions with raw command output, ensuring that every commit or status update is backed by verifiable, real-time data.

主な機能

Automated validation of agent delegation reports against actual VCS diffs
Detection of 'red flag' language such as 'should work' or 'probably fixed'
Rationalization prevention logic to maintain high code integrity standards
Standardized verification patterns for tests, builds, and regression cycles
Mandatory gate function requiring fresh command execution before any status claim
101 GitHub stars

ユースケース

Preventing broken code from being committed by enforcing local test passes
Ensuring bug fixes are valid through rigorous Red-Green TDD verification
Verifying complex build configurations and linter rules before PR creation

概要

主な機能

Automated validation of agent delegation reports against actual VCS diffs
Detection of 'red flag' language such as 'should work' or 'probably fixed'
Rationalization prevention logic to maintain high code integrity standards
Standardized verification patterns for tests, builds, and regression cycles
Mandatory gate function requiring fresh command execution before any status claim
101 GitHub stars

ユースケース

Preventing broken code from being committed by enforcing local test passes
Ensuring bug fixes are valid through rigorous Red-Green TDD verification
Verifying complex build configurations and linter rules before PR creation