claude

Adversarial Review

A reviewer that assumes your code is wrong, in a context window that never saw you write it. Split-context review modelled on Bun's Zig-to-Rust rewrite.

0 on GitHubby Adrian Jonesv0.2.1MITupdated yesterdaydevelopment
/

Plugin adversarial-review from marketplace adrianj98-skills in adrianj98/Skills. Commands read from the repo's manifest — yesterday.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (1)

  • adversary

    Turn adversarial review on or off, check whether it's active, or adversarially review the current diff now. Use when the user asks to enable, disable, pause, or check adversarial review, or asks for an adversarial review of their changes.

Subagents (1)

  • adversary

    Adversarial code reviewer. Reads a diff looking for concrete ways it breaks. Read-only, never edits, never implements. One focused pass, not an exhaustive audit.

Hooks (2)

  • SessionStart

    bash

  • Stop

    bash

View source on GitHub

More from adrianj98/Skills

All 11 plugins →

Caveman

claude

Light caveman mode: short, plain, filler-free replies for the rest of the session. Code, errors and technical terms stay exact; warnings stay in full sentences.

1 skill

0

Deslop

claude

Finds the AI slop in what just changed — restating comments, defensive checks on trusted values, dead abstraction, inflated prose — reports it, and on request cuts it.

1 skill

0

Diag

claude

Pulls an AlertD server's diagnostics by its short name with alertd-core's npm run alertd-server, remembers each name's URL after the first time, and with only a name works out what is wrong from the diagnostics alone.

1 skill

0

Doubt

claude

Questions an analysis or diagnosis the AI gave: splits it into claims, checks the load-bearing ones in a fresh context, names the explanations it skipped, and says what holds, what's wrong and what was never verified.

1 skill

0

Greenlight

claude

Says what it will do and lists the workflows it will watch, then merges the default branch into the current branch, pushes, reruns the PR's required integration check when it was skipped, and fixes the code when a check fails. A skill only, no script.

1 skill

0

Progress

claude

Where a branch and its pull request stand: what it's about, then its state — review, checks, conflicts, what's unpushed, and the one thing left before it merges.

1 skill

0

Similar plugins

mattpocock-skills

claude

Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).

39 skills

mattpocock/skills 323,108

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

cli-anything

claude

Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.

5 commands

HKUDS/CLI-Anything 58,665

agent-browser

claude

Automates browser interactions for web testing, form filling, screenshots, and data extraction

12 skills

vercel-labs/agent-browser 50,373

Open Code Review

codex

Use Alibaba Open Code Review from local Codex.

2 skills

alibaba/open-code-review 50,068

Maintain Adversarial Review?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins11 plugins