skill-verify

Enforce evidence-based verification before claiming software tasks are complete.

Updated Jun 16, 2026
One-click install
npx skills add https://github.com/burgebj/claudeoctopus --skill skill-verify-burgebj
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-verify
Source: https://github.com/burgebj/claudeoctopus/tree/main/skills/skill-verify
Command: npx skills add https://github.com/burgebj/claudeoctopus --skill skill-verify-burgebj

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents unsupported completion claims by enforcing evidence-based verification before reporting that work is finished, fixed, or passing.

Core Features & Use Cases

  • Verification Gates: Requires identifying, running, reading, and validating proof commands before making success claims.
  • Quality Assurance Checks: Distinguishes real evidence from assumptions for tests, builds, bug fixes, requirements, and multi-agent workflows.
  • Use Case: Use this Skill when reviewing code changes or finishing a development task to ensure every reported result is backed by current command output.

Quick Start

Ask the skill to verify that the current implementation is complete and provide evidence before making any success claim.

Frequently Asked Questions about skill-verify

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

FAQPage Schema
How do I enforce evidence-based verification before claiming a code task is complete?▼

Evidence-based verification requires running fresh command execution and validating output before reporting completion. This enforces verification gates to distinguish real proof from assumptions for builds, tests, and bug fixes.

What is the best way to prevent unsupported completion claims during software delivery?▼

To prevent unsupported completion claims, you must identify, run, read, and validate proof commands before making success claims. This ensures every reported result is backed by current command output rather than assumptions.

How do I set up verification gates for multi-provider agent operations and CI workflows?▼

Set up verification gates by requiring fresh command execution and output validation across all development workflows, including multi-provider agent operations and CI, to support reliable completion reporting before tasks are marked passing.

Does this evidence-based verification approach work for code reviews and bug fixes?▼

Yes, this verification approach applies to code reviews and bug fixes by distinguishing real evidence from assumptions. It requires running proof commands and reading their output to validate that the reported fixes actually pass.

When do I need to run fresh command execution for quality assurance checks?▼

You need to run fresh command execution for quality assurance checks whenever you are finishing a development task, reviewing code changes, or completing a delivery phase to ensure every success claim is backed by current command output.