How do I install the Quality skill?

Run npx skillmds@latest add persimmon-automation-labs/quality in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

What does the Quality skill do?

Quality — Index It is listed under Coding & Dev Tools on SkillMD.

Is Quality safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Quality?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Quality free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Quality?

Persimmon-Automation-Labs (@persimmon-automation-labs) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Quality by persimmon-automation-labs

Name: Quality
Rating: 4.2 (5 reviews)
Author: Persimmon-Automation-Labs

File contents

Quality — Index

Persimmon quality = a set of focused review passes plus an orchestrator that runs them before delivery. This mother is a map; follow the child for the actual work.

Trigger

"Review this change" / "pre-delivery QA" / "ship check"
"Is this fast enough / type-safe / DB-efficient?"
"Final review before the client sees it"

The child skills

Skill	When to use	Owns
`quality-final-review`	Before any client delivery	Orchestrates all review-* dimensions + `security-review`; produces a go/no-go summary
`quality-review-performance`	Perf concerns	RSC boundaries, bundle size, DB query plans, Core Web Vitals
`quality-review-type-safety`	Type rigor	Strict-mode gaps, `any` leaks, unchecked casts, missing Zod
`quality-review-data-layer`	DB review	N+1 queries, unbounded lists, missing indexes, transaction boundaries
`quality-review-prompt-output`	AI output review	Prompt hygiene, output validation, citation grounding, eval harness

How to route

Delivering? → quality-final-review (it pulls in the others + security-review).
Single dimension? → invoke the matching quality-review-* child directly.

Persimmon quality defaults — one-screen summary

Final review is mandatory before client delivery — never skip.
Review dimensions: performance, type-safety, data-layer, prompt-output, security.
Type-safety: no any without justification; Zod at every boundary.
Data-layer: no N+1, no unbounded lists, indexes present, transactions where needed.
AI output: every claim grounded/cited where the domain requires it (e.g. legal briefs).
Verification is evidence-based: run tsc --noEmit, eslint, prisma validate, tests — don't assert "looks fine".

Anti-patterns banned

Shipping to a client without quality-final-review
Declaring a review passed without running the checks
Ignoring N+1 / unbounded lists "for now"
Accepting ungrounded AI output where citations are required

Relationship to other mothers

Mother	Connection
`workflow`	`workflow-verify` and `workflow-code-review` invoke these children
`security`	`security-review` is folded into `quality-final-review`
`stack` / `data` / `ai`	Each review dimension audits the matching domain

---name: qualitydescription: Quality — Index---# Quality — IndexPersimmon quality = a set of focused review passes plus an orchestrator that runs them before delivery. This mother is a map; follow the child for the actual work.## Trigger- "Review this change" / "pre-delivery QA" / "ship check"- "Is this fast enough / type-safe / DB-efficient?"- "Final review before the client sees it"## The child skills| Skill | When to use | Owns ||---|---|---|| `quality-final-review` | Before any client delivery | Orchestrates all review-* dimensions + `security-review`; produces a go/no-go summary || `quality-review-performance` | Perf concerns | RSC boundaries, bundle size, DB query plans, Core Web Vitals || `quality-review-type-safety` | Type rigor | Strict-mode gaps, `any` leaks, unchecked casts, missing Zod || `quality-review-data-layer` | DB review | N+1 queries, unbounded lists, missing indexes, transaction boundaries || `quality-review-prompt-output` | AI output review | Prompt hygiene, output validation, citation grounding, eval harness |## How to route1. **Delivering?** → `quality-final-review` (it pulls in the others + `security-review`).2. **Single dimension?** → invoke the matching `quality-review-*` child directly.## Persimmon quality defaults — one-screen summary- **Final review is mandatory before client delivery** — never skip.- **Review dimensions**: performance, type-safety, data-layer, prompt-output, security.- **Type-safety**: no `any` without justification; Zod at every boundary.- **Data-layer**: no N+1, no unbounded lists, indexes present, transactions where needed.- **AI output**: every claim grounded/cited where the domain requires it (e.g. legal briefs).- **Verification is evidence-based**: run `tsc --noEmit`, `eslint`, `prisma validate`, tests — don't assert "looks fine".## Anti-patterns banned- Shipping to a client without `quality-final-review`- Declaring a review passed without running the checks- Ignoring N+1 / unbounded lists "for now"- Accepting ungrounded AI output where citations are required## Relationship to other mothers| Mother | Connection ||---|---|| `workflow` | `workflow-verify` and `workflow-code-review` invoke these children || `security` | `security-review` is folded into `quality-final-review` || `stack` / `data` / `ai` | Each review dimension audits the matching domain |

Frequently asked questions

Quality