verify

Run full verification commands and match evidence to completion claims.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/CasaPerks/casaflow --skill verify-casaperks
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: verify
Source: https://github.com/CasaPerks/casaflow/tree/main/core/skills/verify
Command: npx skills add https://github.com/CasaPerks/casaflow --skill verify-casaperks

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents false “we’re done / it passes / it’s fixed” claims by requiring fresh verification evidence immediately before any completion, success, or satisfaction statements.

Core Features & Use Cases

  • Verification gate for completion claims: Forces you to identify the exact command that proves your claim, run the full command, and only then state the result with evidence.
  • Evidence-first reporting: Blocks vague or synonymous success language unless it is supported by the specific verification output you just produced.
  • Feature-flag specific verification: When a feature flag decision is enabled in the spec, requires three distinct evidence artifacts (ON behavior, OFF behavior, and a matching PostHog event link) before concluding verification is complete.

Quick Start

Use the verify skill right before you commit, create a PR, or report status, by running the exact verification command that proves your success claim and pasting the full output back into the chat.

Frequently Asked Questions about verify

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

FAQPage Schema
How do I enforce fresh verification evidence before claiming a pull request is ready?

Fresh verification evidence requires running the exact verification command in-context immediately before any PR readiness claim, then matching the produced output directly to the specific success statement. This prevents false completion claims by blocking vague success language unless supported by the freshly executed command output.

What is feature flag verification and when do I need ON, OFF, and PostHog event evidence?

Feature flag verification requires three distinct evidence artifacts—ON behavior, OFF behavior, and a matching PostHog event link—before concluding verification is complete. This specific process is needed when a feature flag decision is enabled in your spec frontmatter and gates your release readiness.

How to stop false success claims during CI/CD release gating and commit preparation?

To stop false success claims during CI/CD release gating, you must identify the exact command proving your claim, run the full command fresh in-context, and only then state the result with the pasted evidence. This evidence-first reporting blocks synonymous success language without specific verification output.

Does verification evidence logging work without external dependencies for PR status updates?

Yes, evidence logging works without external dependencies by requiring you to run the full verification command directly in your context and paste the output back into the chat. It operates as a standalone gate for task status updates, work delegation, and commit preparation without needing additional tools.

Best way to verify task completion when reporting test or build passes?

The best way to verify task completion is applying an evidence-first gate that requires fresh execution of the full verification command immediately before reporting that tests or builds have passed. You must match the fresh output to the specific claim, ensuring no stale or assumed results are reported.

Why does my completion claim get blocked if I don't run the verification command again?

Your completion claim gets blocked because evidence-first verification requires fresh execution of the full command in-context immediately before any success statement. It prevents false claims by rejecting vague or synonymous success language that lacks the specific, freshly produced verification output matching the claim.