verify-before-done

Require explicit evidence before claiming task completion.

1|Updated Jan 31, 2026
One-click install
npx skills add https://github.com/kjaylee/misskim-skills --skill verify-before-done
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: verify-before-done
Source: https://github.com/kjaylee/misskim-skills/tree/main/skills/verify-before-done
Command: npx skills add https://github.com/kjaylee/misskim-skills --skill verify-before-done

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents premature claims of task completion by enforcing a strict verification process before any completion status is announced or acted upon.

Core Features & Use Cases

  • Mandatory Verification: Requires concrete evidence for any completion claim (e.g., tests passing, builds succeeding, bugs fixed).
  • Prevents False Positives: Stops the AI from reporting work as done when it's not, reducing errors and rework.
  • Use Case: Before reporting a bug fix as complete, the AI must re-run the specific test case that previously failed and present the passing output as proof.

Quick Start

Use the verify-before-done skill to ensure all tests have passed before claiming the current task is complete.

Frequently Asked Questions about verify-before-done

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

FAQPage Schema
How do I prevent my AI agent from claiming a bug fix is complete before tests pass?

Task completion verification works by mandating concrete evidence, such as passing test outputs or successful build reports, for every completion assertion to prevent false positives and ensure accuracy across all development stages.

What is the best way to enforce validation before a code commit is marked as done?

The best way to enforce validation before a code commit is to apply a rigorous verification protocol that requires explicit, fresh evidence of successful builds or passing tests before any completion status is announced or acted upon.

How do I stop false positive work reports during software development?

Stop false positive work reports by requiring mandatory verification that presents concrete evidence for any completion claim, which prevents the AI from reporting work as done when it is not and reduces rework.

Does this verification process apply to all stages of task management and development?

Yes, this verification process applies to all stages of development and task management, including code commits, build reports, and bug fixes, ensuring rigorous validation before any task completion is claimed.

What limitations exist when relying on AI for quality assurance without explicit evidence?

Without explicit evidence, quality assurance limitations include false claims of task completion and increased rework, because the AI may report work as done prematurely if strict validation requiring fresh proof is not enforced.