aat

Formalize acceptance testing by generating hypotheses and evidence bundles for PR changes.

1|Updated Jan 22, 2026
One-click install
npx skills add https://github.com/guardian-intelligence/apm2 --skill aat
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aat
Source: https://github.com/guardian-intelligence/apm2/tree/main/documents/skills/aat
Command: npx skills add https://github.com/guardian-intelligence/apm2 --skill aat

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill standardizes the PR acceptance process by enabling teams to specify hypotheses, perform structured verifications, and prevent merges that do not meet predefined criteria.

Core Features & Use Cases

  • Hypothesis-driven verification: Formulates testable predictions about PR changes and validates them before merge.
  • Evidence-driven decisions: Produces an evidence bundle detailing outcomes, timestamps, and results for auditability.
  • Anti-gaming checks: Includes static analysis for inconsistencies, hardcoded values, or hidden pitfalls to ensure integrity.
  • Use Case: When evaluating a large PR, the AAT protocol guides the tester to craft hypotheses, execute verification steps, and generate a reproducible evidence report to support the decision.

Quick Start

Activate the AAT protocol against a pull request, define hypotheses, run verification on the PR changes, and generate the evidence bundle to guide the merge decision.

Frequently Asked Questions about aat

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

FAQPage Schema
How do I verify pull request changes before merging using acceptance testing?

Acceptance testing for pull request verification involves creating testable hypotheses about PR changes, executing structured validation steps, and generating an evidence bundle to support the merge decision. This ensures changes meet predefined criteria under varying inputs.

What is hypothesis-driven PR validation and how does it work?

Hypothesis-driven PR validation is a structured acceptance testing method where you formulate testable predictions about code changes, run deterministic verification against those hypotheses, and produce an evidence bundle of outcomes for auditability and QA governance.

How do I prevent hardcoded values or hidden pitfalls from passing PR verification?

To prevent hardcoded values from passing PR verification, apply anti-gaming static analysis checks during acceptance testing. These checks detect inconsistencies, hidden pitfalls, and artificial test pass conditions to ensure PR integrity before merging.

How do I generate an evidence bundle for QA governance during PR review?

Generate an evidence bundle for QA governance by running structured hypothesis verification on PR changes, which automatically captures outcomes, timestamps, and deterministic test results into a reproducible audit report for merge decision support.

Does hypothesis-driven acceptance testing work for large pull requests?

Hypothesis-driven acceptance testing is highly effective for large pull requests. The protocol guides testers to systematically craft hypotheses, execute verification steps across varying inputs, and generate a reproducible evidence report to manage complexity.

What is the best way to standardize PR acceptance criteria across a development team?

The best way to standardize PR acceptance criteria is implementing a hypothesis-driven validation protocol that enforces structured verification, automated evidence-bundle production, and anti-gaming checks aligned with your team's QA governance standards.