gsd-verify-work

Run conversational UAT per phase and generate {phase_num}-UAT.md reports.

1|Updated Sep 6, 2015
One-click install
npx skills add https://github.com/deerawan/dotfiles --skill gsd-verify-work-deerawan
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: gsd-verify-work
Source: https://github.com/deerawan/dotfiles/tree/main/claude/skills/gsd-verify-work
Command: npx skills add https://github.com/deerawan/dotfiles --skill gsd-verify-work-deerawan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

gsd-verify-work helps you avoid shipping broken or misunderstood features by validating the outcome from the user’s perspective through conversational UAT, then guiding diagnosis and fix planning when issues appear.

Core Features & Use Cases

  • Conversational UAT with persistent tracking: Runs one test at a time and records results into a phase-based report so you can see exactly what passed or failed.
  • Issue diagnosis and fix-plan preparation: When something doesn’t work, it identifies the gaps and prepares verified fix plans instead of stopping at a failure.
  • Workflow-safe routing to execution: Preserves existing workflow gates and prepares for a later /gsd:execute-phase step only when appropriate.

Example use case: you want to validate that a feature built by an AI (across multiple phases) truly satisfies end-user expectations before moving to execution.

Quick Start

Run gsd-verify-work for a specific phase by providing the phase number argument, for example: "5".

Frequently Asked Questions about gsd-verify-work

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

FAQPage Schema
How do I run conversational UAT to validate features from a user perspective?▼

Conversational UAT validates features by running one test at a time from the user perspective, tracking results per phase in a report to confirm what was built actually works.

What is the best way to diagnose issues and prepare fix plans during user acceptance testing?▼

Issue diagnosis during user acceptance testing identifies gaps when features fail and prepares verified fix plans instead of stopping, ensuring routing to execution is workflow-safe.

How do I verify multi-phase work without breaking existing workflow gating?▼

Verifying multi-phase work requires session-aware workflow gating that detects active sessions, runs end-to-end execution, and preserves workflow gates for later execution phases.

Can I test a specific phase of an iterative build process for fix-plan readiness?▼

Yes, you can test a specific phase of an iterative build process by providing the phase number, which generates a UAT report to assess fix-plan readiness for that phase.

Why does feature validation require persistent tracking during end-to-end testing?▼

Feature validation needs persistent tracking because multi-phase work requires recording pass or fail results per phase into a report, ensuring accurate diagnosis and routing to execution.