Is the Production Grade skill free?

Yes. The skill itself is free from a-tokyo/agent-skills. SkillProof publishes the install command and an independent test verdict at no cost.

What is the Production Grade SkillProof Score?

9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.

How do I install Production Grade?

Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.

Can I use Production Grade with Cursor, Copilot, Gemini CLI, Codex or other AI tools?

The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.

Production Grade

A 16-rule senior-engineer posture: plan first, atomic writes, typed errors, schema+queries+indexes as one diff.

by a-tokyo · a-tokyo/agent-skills

Tested · Works

Install — copy the command Download source ↗

Test report

Verdict: Tested · Works
Score: 9.2/10
Tested: Jul 15, 2026
Environment: Claude Code 2.x (agent harness)
Upstream re-checked: Aug 10, 2026 · cb87fd6

On a login-tracking task the bare-model baseline shipped a check-then-act read-then-update with a lost-update race under concurrent logins; following the skill's R5/R6/R9 rules produced an atomic single-statement UPDATE, a migration with a down-migration, and a concurrency test — a concrete, verifiable fix to a real bug the baseline introduced. The repo's own README ships a self-run benchmark (idempotency 0%->80-90% across three models) and is refreshingly honest about where the skill underperforms (a vague rate-limit ask, one Opus regression cell) rather than only showing wins.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

Installs cleanly 5/5
Triggers reliably 5/5
Output vs. baseline 8/10
Docs & honesty 5/5

What Production Grade does

Loads a dense checklist of production-engineering rules (idempotent writes, N+1 avoidance, schema/migration hygiene, typed errors, security-by-plan, test-first) plus a mandatory self-verification gate run against every diff before submission. Triggers on non-trivial planning, implementation, review, refactors, RCA, or hardening inherited/LLM-generated code; explicitly skips one-line typo fixes.

How to install Production Grade

git clone https://github.com/a-tokyo/agent-skills
mkdir -p ~/.claude/skills
cp -r agent-skills/skills/production-grade ~/.claude/skills/production-grade

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Production Grade

/production-grade A 16-rule senior-engineer posture: plan first, atomic writes, typed errors, schema+queries+indexes as one diff.

It also activates on plain-language prompts like these:

Implement this properly, no shortcuts, make it solid for production.
This code was vibe-coded, can you harden it for real production use?
Add a login-count increment endpoint without any race conditions.

Frequently asked questions

Is the Production Grade skill free?: Yes. The skill itself is free from a-tokyo/agent-skills. SkillProof publishes the install command and an independent test verdict at no cost.
Does Production Grade work with Claude Code?: We tested it with Claude Code 2.x (agent harness) on Jul 15, 2026. Verdict: Tested · Works. On a login-tracking task the bare-model baseline shipped a check-then-act read-then-update with a lost-update race under concurrent logins; following the skill's R5/R6/R9 rules produced an atomic single-statement UPDATE, a migration with a down-migration, and a concurrency test — a concrete, verifiable fix to a real bug the baseline introduced. The repo's own README ships a self-run benchmark (idempotency 0%->80-90% across three models) and is refreshingly honest about where the skill underperforms (a vague rate-limit ask, one Opus regression cell) rather than only showing wins.
What is the Production Grade SkillProof Score?: 9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.
How do I install Production Grade?: Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Production Grade with Cursor, Copilot, Gemini CLI, Codex or other AI tools?: The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.