Adversarial Self Review — Development skill for Claude Code
Skill: Adversarial self-review — try to break your own work before shipping.
How to install Adversarial Self Review
Installs to ~/.claude/skills/contact715-jidoka-adversarial-self-review/SKILL.md
mkdir -p ~/.claude/skills/contact715-jidoka-adversarial-self-review && curl -fsSL https://raw.githubusercontent.com/contact715/jidoka/HEAD/.claude/skills/adversarial-self-review.md -o ~/.claude/skills/contact715-jidoka-adversarial-self-review/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Adversarial Self Review does
Skill: Adversarial self-review — try to break your own work before shipping
Wave: w28 | Status: experimental | Tags: [process, qa, falsification]
When to use
- Before any commit that contains > 30 LOC of new code.
- Before declaring a wave done.
- Before responding "this is fixed" to a user.
The instinct after finishing a piece of work is to **verify** it (does it pass tests? does TS compile?). Verification finds known failure modes. Adversarial review finds **unknown** failu
Alternatives in Development
- Om Tidy — Self-maintenance pass — acts on every hygiene flag: archives completed work, groups loose clusters, splits ove 4.6k ★
- Defensive Code Cleanup — Clean Defensive Try-Catch Blocks 1.2k ★
- AgentHandover — What if OpenClaw, Claude Code, Codex etc. knew how to do your work without you saying it 600 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Payne Review
Run a PayneSDD adversarial (break-it) review of the current changes
Mem Save
Save a curated session-level summary to history. Use this at meaningful milestones — end of session, before a
Deep Review
Run a fresh-context, un-anchored ADVERSARIAL-LOGIC review of the current diff before
Adversarial Plan
You are an independent reviewer performing an adversarial review of an implementation plan before any code is
Feature Forge
A full feature lifecycle for coding agents: crucible, brief, wireframe, architecture, plan, TDD build loop, ad
Builder Self Review
Builder step 5 — re-read your own diff before publishing
Related Agents
Lw Verifier
Adversarial verification. Use to independently check that a change is correct, that tests genuinely pass, or t
Antagonistic Reviewer
Adversarial single-lens reviewer invoked by ship- skills (design, create-plan) to try to break an artifact thr
Slop Detector
Adversarial generic-design hunt with fresh context — finds structural, component, visual, copy, interaction, a
