review

A read-only, two-pass code review method focused on security, consistent interfaces, and how changes affect users.

In plain words
What is it for?
Use it when reviewing a change, especially for security issues, mismatched APIs, or confusing user-facing behavior.
Why use it?
It first searches aggressively for possible problems, then checks those findings again to reduce false alarms and unsupported conclusions.

Skill for Claude CodeCodex

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/lestrrat-ai/claude-code-plugins/review
Any agent
npx skills add lestrrat-ai/claude-code-plugins --skill review
Clone the repo
git clone --depth 1 https://github.com/lestrrat-ai/claude-code-plugins

Made for: Claude Code, Codex.

Per session 156 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 7,499 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 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 $0.00156 $0.07499
Opus 5 $0.00078 $0.03750
Sonnet 5 $0.00031 $0.01500
Haiku 4.5 $0.00016 $0.00750

Measured 2d ago against content hash 7db9388f4597, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

review scanned grade B 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 2d 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.

Recursive force deletemediumDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

Before starting, `mkdir -p .gauntlet/tmp` and delete stale `.gauntlet/tmp/review-*` files from previous runs. `.gauntlet/` should be git-ignored; if it is **not**, **warn the user** rather than editing `.gitignore` — rep

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

plugins/gauntlet/skills/review/SKILL.md · 476 lines

How it starts

The opening of the file, as written. The whole thing — 476 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Review

Two phases, in order. The hostile pass finds; the neutral pass filters. Skipping phase 2 means delivering noise.

Runtime Adapter

This skill runs in Claude Code and Codex.

  • Resolve <SKILL-PATH> from the active SKILL.md path supplied by the host. Do not depend on CLAUDE_PLUGIN_ROOT, PLUGIN_ROOT, or another plugin-root environment variable.
  • "Subagent" means a fresh, context-isolated worker launched through the host's agent mechanism.
  • Dispatch independent workers in one parallel batch when the host supports it. Give each worker its scope, permissions, and output path in the prompt; do not require a host-specific agent type.
  • Constrain review workers to read-only work. Grant command execution only when an audit must run a verification command.
  • If the host has no agent mechanism, run Phase 1 lens passes sequentially and use the Phase 2 same-context fallback. Report that context isolation was unavailable.
  • Claude Code invocation: /gauntlet:review. Codex invocation: $gauntlet:review.

Phase 1 — Hostile Pass

Treat the code as a suspect, not a colleague. Goal: surface everything that could be wrong.

Mindset

  • Assume every input is malicious, every caller is confused, every assumption is wrong.
  • "It works on happy path" is not evidence. Adversary picks the path.
  • Author's intent does NOT excuse behavior. Only the code's actual behavior matters.
  • No diplomatic softening. State the defect, the impact, the trigger.

Scope

Resolve target before reviewing:

  • PR / branch → git diff <base>...HEAD (and read full files for changed regions, not just hunks)
  • Package / dir → enumerate files, read each
  • Single file → read whole file

Always read enough surrounding context to understand callers and invariants. Hunks alone hide bugs.

Sizing — pick a strategy

Measure surface BEFORE reading code:

  • PR/branch: git diff --stat <base>...HEAD (changed-file count, lines added+removed)
  • Dir/package: enumerate files with the host's file-listing/search tool (count = file count), then wc -l <files> for LOC.

Read the full file on GitHub · 476 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 2d ago First seen · 476 lines · 156 tokens per session scan B 7db9388f4597

Subscribe to this mod's changes

review is a skill published in the GitHub repository lestrrat-ai/claude-code-plugins (18 stars, last pushed 4d ago), licensed MIT. It adds 156 tokens to every session and 7,499 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it B with 1 finding (recursive force delete). 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