When you spin up a dev server or mutter "why isn't this working," this skill kicks in to trace the full user story from browser through API to data and back. It's not running unit tests. It's inferring what you're actually building, then checking each boundary with evidence: screenshots, terminal output, network logs, environment state. The reporting format is structured (what was checked, what was found, what's next), and it stops at the first broken boundary instead of piling on noise. Honestly most useful when you've wired up a feature across layers and need to confirm the whole thing actually works, or when something feels off but you're not sure which layer is lying to you.
npx -y skills add vercel-labs/vercel-plugin --skill verification --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
microsoft/azure-skills
zxkane/aws-skills
awslabs/agent-plugins
microck/ordinary-claude-skills
microsoft/github-copilot-for-azure
zxkane/aws-skills