omk-evidence-contract

Generate task-specific evidence contracts that validate or falsify completion claims.

130|14|Updated Apr 30, 2026
One-click install
npx skills add https://github.com/dmae97/oh-my-kimi --skill omk-evidence-contract
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: omk-evidence-contract
Source: https://github.com/dmae97/oh-my-kimi/tree/main/templates/skills/agents/omk-evidence-contract
Command: npx skills add https://github.com/dmae97/oh-my-kimi --skill omk-evidence-contract

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents premature “done” confirmations by forcing a structured evidence contract that separates verified results, missing gaps, and remaining risk.

Core Features & Use Cases

  • Task-specific completion gating: Turns a one-sentence completion claim into a checklist of minimum proof required to support or falsify it.
  • Scenario-aware evidence types: Supports feature/bugfix/refactor, research, release, and security tasks with tailored evidence categories (diffs, tests, citations, audits, rollback notes).
  • Honest uncertainty capture: Explicitly records unavailable evidence with concrete reasons and lists conflicting evidence without overstating success.

Quick Start

Ask an AI to generate an evidence contract for your upcoming change by providing the task type and your single-sentence completion claim.

Frequently Asked Questions about omk-evidence-contract

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

FAQPage Schema
How do I prevent premature release readiness declarations for code changes?

Generate an evidence contract to validate completion claims before declaring release readiness. It forces a structured checklist separating verified test results from missing gaps and remaining risk, preventing overstated success.

What is an evidence contract and how does it work for security audits?

An evidence contract is a task-specific validation framework for security audits that maps required proof artifacts, explicitly records unavailable evidence with reasons, and provides a final risk assessment without converting gaps into success.

How do I structure code change verification for different task scenarios?

Use scenario-aware evidence types tailored to feature, bugfix, refactor, research, release, and security tasks. Each scenario requires specific proof artifacts like diffs, tests, citations, audits, and rollback notes to verify completion.

How do I capture honest uncertainty and missing test results before declaring work complete?

Record unavailable evidence with concrete reasons and list conflicting evidence without overstating success. The evidence contract explicitly separates verified results from missing gaps to maintain honest uncertainty capture during risk assessment.

Does evidence contract validation work for research and documentation tasks?

Yes, evidence contracts apply to research, docs, and orchestration scenarios alongside code changes. They validate completion claims by mapping required proof artifacts like citations and explicitly flagging uncertainty and risk.

What's the best way to turn a one-sentence completion claim into verifiable proof?

Provide the task type and single-sentence completion claim to generate a task-specific evidence contract. It transforms the claim into a minimum proof checklist that validates or falsifies the work before completion.