verification-before-completion

Name: verification-before-completion
Availability: InStock
Author: bbohhu-prog

Run verification commands and report results only after confirming exit status and outputs.

Updated Apr 12, 2026

Search Tags:#verification #safety #evidence #ai-assistant #claims #gatekeeping #prompt-injection

Ready to run

npx skills add https://github.com/bbohhu-prog/waxs --skill verification-before-completion-bbohhu-prog

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: verification-before-completion
Source: https://github.com/bbohhu-prog/waxs/tree/main/.claude/skills/superpowers/verification-before-completion
Command: npx skills add https://github.com/bbohhu-prog/waxs --skill verification-before-completion-bbohhu-prog

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Enforces verification before making completion claims to prevent dishonest or premature success statements.

Core Features & Use Cases

BEFORE claiming status: IDENTIFY, RUN, READ, VERIFY, and ONLY THEN claim, providing guardrails against premature completion.
Applies to software development, code reviews, PRs, and any scenario where an assertion of success must be backed by fresh outputs.
Reduces risk of false assurances by ensuring outputs and exit statuses are checked and documented.

Quick Start

Run the verification sequence on your latest changes, capture fresh outputs, and report the exact results before making any completion claims.

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 an AI assistant from claiming code changes are complete before they are verified?▼

To prevent premature completion claims, enforce a verification gate that requires running a full verification command and inspecting fresh outputs. Only after checking exit statuses and confirming evidence of success can the AI assistant report results.

How do I verify CI pipeline success using fresh outputs before approving a pull request?▼

Verify CI pipeline success by executing your full verification sequence on the latest changes, capturing fresh outputs, and reading the exact exit status. You report results only when the evidence confirms the pipeline passed.

What is evidence-first verification for software development claims?▼

Evidence-first verification is a gatekeeping process that enforces identifying, running, reading, and verifying commands before making any success claims. It ensures assertions are backed by fresh outputs to reduce the risk of false assurances.

Can I use automated verification gatekeeping for code reviews and PR statuses?▼

Yes, you can apply automated verification gatekeeping to code reviews and PRs. It requires running a full verification command, inspecting the exit status and outputs, and reporting results only when evidence confirms success.

What is the best way to stop false success assertions in AI-assisted coding?▼

The best way to stop false success assertions is to implement guardrails that require running a verification sequence and capturing fresh outputs before claiming status. This prevents dishonest or premature success statements.

Why does my AI assistant claim tasks are done without inspecting the exit status?▼

AI assistants claim tasks are done without inspecting exit status when lacking verification guardrails. Enforcing a process to identify, run, read, and verify outputs before claiming completion reduces the risk of these false assurances.