verify

Run fresh verification commands and inspect outputs before claiming task completion.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/mengdehong/opencode-dotfiles --skill verify-mengdehong
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: verify
Source: https://github.com/mengdehong/opencode-dotfiles/tree/main/skills/verify
Command: npx skills add https://github.com/mengdehong/opencode-dotfiles --skill verify-mengdehong

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prior to claiming work is complete, this skill enforces evidence-backed verification to prevent premature or dishonest completion claims.

Core Features & Use Cases

  • Enforces an explicit verification step before any completion claim, reducing risk of undiscovered regressions or silent failures.
  • Applies to code changes, builds, tests, and reviews, ensuring every claim is grounded in fresh command outputs and observable results.
  • Use Case: Before creating a PR or marking a task done, run the verification commands, inspect outputs, and attach evidence to validate completion.

Quick Start

Run the verification command for the current task and attach its fresh output as evidence before making any completion claim.

Frequently Asked Questions about verify

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

FAQPage Schema
How do I verify code changes before claiming a task is complete?

To verify code changes, identify a specific command that proves the completion claim, run it fresh, inspect the full output, check exit codes, and only then assert success.

What is evidence-backed verification in software engineering?

Evidence-backed verification is the process of requiring fresh command outputs and observable results before asserting that a coding task, build, or review is complete.

How do I prevent premature completion claims when finishing a coding task?

Prevent premature completion claims by enforcing an explicit verification step that requires running a relevant command, reading its output, and attaching fresh evidence.

Can I use automated verification for build and test completion workflows?

Yes, automated verification applies to builds, tests, and code reviews by requiring you to run the appropriate verification commands and examine their outputs before marking completion.

Does verification require checking exit codes and reading full command output?

Yes, verification requires reading the full output of the executed command and checking its exit codes to ensure the completion claim is grounded in observable results.

What is the best way to validate a pull request before marking it done?

The best way to validate a pull request is to run the relevant verification commands, inspect their fresh outputs for regressions or failures, and attach that evidence before completion.