codex-adversarial

A review skill that deliberately challenges a code change by looking for security, correctness, and other reasons not to ship it. This kind of review is often called adversarial or red-team testing.

In plain words
What is it for?
Use it when you want a skeptical review of a change before release, with findings saved to a report.
Why use it?
It helps uncover problems that a normal, confirmation-focused code review may miss.

Skill for Claude CodeCodex

Part of the codex-advisor plugin — 10 skills shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/leejuoh/claude-code-zero/codex-adversarial
Any agent
npx skills add LeeJuOh/claude-code-zero --skill codex-adversarial
Clone the repo
git clone --depth 1 https://github.com/LeeJuOh/claude-code-zero

Made for: Claude Code, Codex.

Or install codex-advisor, the plugin that ships this one along with the rest of its 10 skills.

Per session 46 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,860 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00046 $0.02860
Opus 5 $0.00023 $0.01430
Sonnet 5 $0.00009 $0.00572
Haiku 4.5 $0.00005 $0.00286

Measured 3d ago against content hash 0600df6cae08, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

codex-adversarial scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Reads agent configuration directorieslowAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

`--model` and `--effort` route through `scripts/apply-codex-config.py` to update `~/.codex/config.toml` before the companion launches. Two reasons:

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

plugins/codex-advisor/skills/codex-adversarial/SKILL.md · 260 lines

How it starts

The opening of the file, as written. The whole thing — 260 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Codex Adversarial Review + Double-Check

You are a translator + executor + double-checker. Adversarial review defaults to skepticism — it looks for reasons NOT to ship. Because adversarial hallucinates more than plain review, Phase 4 rigor matters even more than usual.

Execution Contract

This contract overrides default exploration habits. Read it before Phase 1.

Phase Allowed Forbidden
1 ANALYZE test -f/-s/-d, git rev-parse --verify, git branch --list, wc -l/-c, file, echo, printf cat, head, tail, git diff, git log -p, git show, git blame, Read, Grep, Glob
2 INVOKE Bash for companion launch (multi-arg form only — never $ARGUMENTS blob) All source reads
3 WAIT BashOutput All source reads, manual polling, ps/kill outside KillShell
4 DOUBLE-CHECK Read ONLY files/lines Codex cited Reading whole files "for context"; reading uncited files; inventing citations
5 REPORT + SAVE Write report file n/a

The companion collects the diff itself. Unknown flags are silently joined into the prompt by the companion (lib/args.mjs:47-49 + :643-650). Phase 1 whitelist is the only safety net.


Phase 1: Analyze

You are a translator. Use LM intelligence, not regex tables.

Whitelist for this skill: --base <ref>, --scope <auto|working-tree|branch>, --model <slug>, --effort <level>, and positional focus text (natural-language attack hints, e.g., "check for SQL injection in login handler").

--model and --effort route through scripts/apply-codex-config.py to update ~/.codex/config.toml before the companion launches. Two reasons:

  1. --effort is not a registered review flag (handleReviewCommand valueOptions = ["base", "scope", "model", "cwd"] at :714). Passing --effort directly would become silent prompt corruption (references/companion-usage.md §3). Only the config.toml model_reasoning_effort key reaches the review path.
  2. Consistency + persistence. --model IS honored as a flag in v1.0.4+ (startThread({ model }), lib/codex.mjs:1010-1015), but routing it through config.toml keeps every codex-advisor skill identical and lets the value persist for the next session without re-typing.

Read the full file on GitHub · 260 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 260 lines · 46 tokens per session scan A 0600df6cae08

Subscribe to this mod's changes

codex-adversarial is a skill published in the GitHub repository LeeJuOh/claude-code-zero (51 stars, last pushed 3d ago), licensed MIT. It adds 46 tokens to every session and 2,860 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens

chat-perf

Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.

microsoft/vscode · 51 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens