Is the Review Discipline skill free?

Yes. The skill itself is free from Skillproofdev/review-discipline. SkillProof publishes the install command and an independent test verdict at no cost.

What is the Review Discipline 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 Review Discipline?

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 Review Discipline 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.

Review Discipline

Findings-only code review: hunt wide, prove every bug, no praise filler.

by SkillProof · Skillproofdev/review-discipline

Tested · Works

Install — copy the command Download source ↗

Test report

Verdict: Tested · Works
Score: 9.2/10
Tested: Jul 11, 2026
Environment: Claude Code 2.x (agent harness)
Upstream re-checked: Aug 10, 2026 · dfeda87

Our own skill — tested under the same protocol, disclosed. Seeded-bugs bench (4 samples, 21 seeded bugs, 12 FP traps): the skill (v2) catches 18.5/21 bugs (88%) vs 17/21 (81%) unguided, goes 7/7 on P2 edge-case bugs vs base's 6/7, and holds 0% false positives vs base's 5.6%, with ~100% of findings carrying a demonstrated failure path vs ~63% for base. One disclosed caveat: v2 is not a strict superset of the skill's own v1 — it traded one subtle deep-nesting P3 catch for the broader P2 sweep. Protocol test caught real issues in our own repos (dead benchmark links, strict-YAML frontmatter) — fixed and re-verified same day.

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 Review Discipline does

Enforces a strict review order — read the full diff and its context, hunt exhaustively for anomalies before verifying any of them, then attach a demonstrated failure path (or an explicit [suspicion] tag) to every finding. Outputs P0–P3 ranked findings only, with no praise filler and no undemonstrated claims asserted as fact. Stays out of feature work, style-only linting, and prose review by design.

Review Discipline before / after — same task, with and without

Without the skill

Unguided review of a seeded-bug sample: 17/21 bugs caught (81%), 1 false positive, only ~63% of findings backed by a demonstrated failure path, and 3 of 7 edge-case (P2) bugs missed — including rounding/truncation errors in billing code.

With Review Discipline

With review-discipline's breadth-before-depth hunt: 18.5/21 bugs caught (88%), a clean 7/7 on P2 edge cases, 0 false positives, and ~100% of findings carry a demonstrated failure path plus explicit [suspicion] tags where one can't be built — though the exhaustive-breadth v2 rewrite trades away one subtle deep-nesting catch that an earlier v1 uniquely found.

How to install Review Discipline

git clone https://github.com/Skillproofdev/review-discipline ~/.claude/skills/review-discipline

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

Commands — how to trigger Review Discipline

/review-discipline Findings-only code review: hunt wide, prove every bug, no praise filler.

It also activates on plain-language prompts like these:

Review this PR before I merge it — anything I'm missing?
Find bugs in this diff, especially around the billing changes.
Can you check this branch for problems before we ship?

Frequently asked questions

Is the Review Discipline skill free?: Yes. The skill itself is free from Skillproofdev/review-discipline. SkillProof publishes the install command and an independent test verdict at no cost.
Does Review Discipline work with Claude Code?: We tested it with Claude Code 2.x (agent harness) on Jul 11, 2026. Verdict: Tested · Works. Our own skill — tested under the same protocol, disclosed. Seeded-bugs bench (4 samples, 21 seeded bugs, 12 FP traps): the skill (v2) catches 18.5/21 bugs (88%) vs 17/21 (81%) unguided, goes 7/7 on P2 edge-case bugs vs base's 6/7, and holds 0% false positives vs base's 5.6%, with ~100% of findings carrying a demonstrated failure path vs ~63% for base. One disclosed caveat: v2 is not a strict superset of the skill's own v1 — it traded one subtle deep-nesting P3 catch for the broader P2 sweep. Protocol test caught real issues in our own repos (dead benchmark links, strict-YAML frontmatter) — fixed and re-verified same day.
What is the Review Discipline 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 Review Discipline?: 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 Review Discipline 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.