verification-before-completion

Require fresh verification evidence before reporting task completion.

Updated Apr 3, 2026
One-click install
npx skills add https://github.com/norrbacka/threejs_codex_base --skill verification-before-completion-norrbacka
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: verification-before-completion
Source: https://github.com/norrbacka/threejs_codex_base/tree/main/.codex/superpowers/skills/verification-before-completion
Command: npx skills add https://github.com/norrbacka/threejs_codex_base --skill verification-before-completion-norrbacka

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents unverified completion claims by forcing fresh evidence before anyone says a task is done, fixed, or passing.

Core Features & Use Cases

  • Requires running the exact verification command for the current state of the work.
  • Checks output, exit codes, and failure counts before success is reported.
  • Fits testing, debugging, build validation, commit readiness, and PR preparation workflows.
  • Helps teams avoid false confidence, missed regressions, and incomplete handoffs.

Quick Start

Use this skill when you are about to say a task is complete so it can verify the result before you make that claim.

Frequently Asked Questions about verification-before-completion

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

FAQPage Schema
How do I prevent unverified completion claims in my release workflow?

To prevent unverified completion claims, require executing the full verification command and checking exit codes or failure counts before reporting a task as done, fixed, or passing.

What is the best way to verify code review changes before marking a task as done?

The best way to verify code review changes is to run the exact verification command for the current state, read the output and exit codes, and cite that fresh evidence before stating success.

Does this verification approach work with debugging and build validation checks?

Yes, this verification approach works with debugging, build validation, commit readiness, and PR preparation workflows by forcing results to be confirmed in the current run before any success is reported.

Can I use this to require fresh evidence before reporting testing or debugging results?

You can use this to require fresh evidence by executing the full testing or debugging verification command, reading failure counts, and citing the current run's output before claiming a fix is passing.

Why should I require fresh evidence before reporting build checks as passing?

Requiring fresh evidence for build checks prevents false confidence, missed regressions, and incomplete handoffs by ensuring the verification command output and exit codes are confirmed in the current run before success is stated.