skills.rest
gmackie
gmackieCOMMUNITY
Explore all skills by this author →
Website𝕏@grahammackieGitHub

create-verification-skill

Generates a project-local verification skill that drives an app like a user and captures evidence.

Updated Mar 27, 2026
Search Tags:#automation#testing#e2e#skill generation#verification#feature map#Cursor
One-click install
npx skills add https://github.com/gmackie/agent-skills --skill create-verification-skill-gmackie
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: create-verification-skill
Source: https://github.com/gmackie/agent-skills/tree/main/skills/create-verification-skill
Command: npx skills add https://github.com/gmackie/agent-skills --skill create-verification-skill-gmackie

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Many projects lack a scripted way to prove that real user-facing behavior works. This Skill interviews a repository and generates a tailored, project-local verification skill that launches the app, drives it the way a user would, and captures concrete evidence of behavior. ## Core Features & Use Cases - Repo Interview: Detects the app's surface (web UI, CLI/TUI, API, desktop, mobile), run commands, driving harnesses (Playwright, CDP, tmux/PTY, HTTP), and observable evidence sources directly from the codebase. - Skill Generation: Writes a complete .cursor/skills/verify-<app>/SKILL.md with Launch, Doctor, Drive, Evidence, Cleanup, and Helpers sections grounded in real selectors and commands. - Feature Map Seeding: Creates a maintained feature map with one file per user-facing feature, following the example structure in references/feature-map-example/. - Self-Proof: Runs the generated skill end to end once (launch, doctor, drive one feature, capture evidence, clean up) before handing it over. - Use Case: Point it at a repo with no scripted UI testing, and it produces a working verification skill plus a feature map that later agents can execute cold. ## Quick Start Ask the agent to create a verification skill for this repository so the app can be launched, driven, and proven with captured evidence.

Frequently Asked Questions about create-verification-skill

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

FAQPage Schema
How do I create a verification skill for my repository?▼

Invoke the skill on your repo and it interviews the codebase to find the app's surface, dev command, and driving harness, then writes `.cursor/skills/verify-<app>/SKILL.md` with launch, doctor, drive, evidence, and cleanup sections. It also seeds a feature map and proves the skill by running it once.

What app types can a generated verification skill drive?▼

It supports web UIs, CLI and TUI programs, desktop and Electron apps, APIs, services, and mobile apps. The generator prefers existing harnesses like Playwright or Cypress specs, then falls back to generic recipes such as browser/CDP, tmux/PTY sessions, or plain HTTP.

Does the generated verification skill work with Playwright or tmux?▼

Yes. The interview step detects existing harnesses first, including Playwright and Cypress specs, expect scripts, PTY helpers, and curl-able endpoints. For CLI and TUI apps it uses isolated tmux or PTY sessions so concurrent runs do not share state.

What evidence does a verification run capture?▼

Evidence includes screenshots, ARIA snapshots, terminal transcripts, response bodies, logs, exit codes, and database state. Proof standards require capturing the user action plus resulting state and verifying side effects like files written or rows inserted.

Why must the generated skill be executed before handover?▼

A generated skill that was never executed is a draft, not a deliverable. Running launch, doctor, one feature drive, evidence capture, and cleanup end to end confirms the instructions work and that proof artifacts survive teardown.

What are the limitations of generated verification skills?▼

The checkout must build and start as-is; a broken base must be fixed or reported first. If two instances cannot run side by side due to shared ports or data, the generated skill must refuse to double-drive a shared instance rather than risk corrupting the user's session.

Related Skills

Explore More Software Engineering Skills→
nixopusnixopus

dev-setup

Automate local development setup with Node.js, Yarn, PostgreSQL, and Redis.

Official
Intermediate
jeremybrasherjeremybrasher

maintain-verification-skill

Audits and corrects a project's verification skill and feature map against live application behavior.

Community
Advanced
gajjalaashok75-UIgajjalaashok75-UI
2

e2e-testing

Builds Playwright-based end-to-end browser test suites for CI/CD pipelines.

Community
Advanced
hospitaljobsinhospitaljobsin
7

e2e-testing-patterns

Provide end-to-end testing patterns for Playwright and Cypress.

Official
Advanced