codex-adversarial-review SKILL.md skillA
codex-adversarial-review from aarontzeng/dev-lead is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add aarontzeng/dev-lead/codex-adversarial-review@git:20260930.9533005Pins git:20260930.9533005 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- aarontzeng/dev-lead
- Path
skills/codex-adversarial-review/SKILL.md- Versions recorded
- 5
- Latest version
- git:20260930.9533005, committed Sep 30, 2026
- First version recorded
- Sep 23, 2026
- Size
- 16,481 bytes
- Repository stars
- 3
- License
- MIT
Run a read-only Codex challenge review of a git diff with `codex exec` and the suite's adversarial framing (the Codex companion's adversarial-review is the fallback), then verify each finding against the code. Use when the user asks for an adversarial or challenge review from Codex/GPT, wants a second opinion on a design or implementation, or needs the review gate after codex-implement.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Codex adversarial review Run the review as `codex exec` in a `read-only` sandbox, with the lead's lens wrapped in the suite's adversarial framing ([`references/adversarial-framing.md`](references/adversarial-framing.md)) and the effort set per run. Since 0.6.28 this is the default: in an equivalence test on three frozen targets with known answers (codex-runtime.md, 2026-09-23) it matched or beat the companion's `adversarial-review` on known-answer recall and false positives, read the base revision in every run where the companion did in fewer than half, and — unlike the companion — takes the effort the roster or triage asks for. The companion stays as the fallback below. This is a challenge review: inspect chosen boundaries, assumptions, failure modes, and trade-offs, not only local defects. ## Before the first run of a session Read **[`references/codex-runtime.md`](references/codex-runtime.md)** (same directory). It holds the family-level mechanics shared with `codex-implement` — companion resolution, the `--background` launcher-output trap, `status` lying about liveness, the lossy-pipe report loss, report recovery from session rollouts, sandbox …
Read the whole file at its exact version.
How to install
mdr add aarontzeng/dev-lead/codex-adversarial-review@sha256:5139246ac78635bcThe line at the top of this page pins the label git:20260930.9533005, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_5qb5xx7dzakfaahx)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 5 versions of this file, between Sep 23, 2026 and Sep 30, 2026, about one every 2 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260930.9533005 latest | 2026-09-30 | 9533005 | 16,481 B | A | view · diff |
| git:20260925.1a5468f | 2026-09-25 | 1a5468f | 16,337 B | A | view · diff |
| git:20260924.e60cfa5 | 2026-09-24 | e60cfa5 | 15,638 B | A | view · diff |
| git:20260923.7c1829e | 2026-09-23 | 7c1829e | 15,563 B | A | view · diff |
| git:20260923.6231e29 | 2026-09-23 | 6231e29 | 14,616 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (16481 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
aarontzeng/dev-lead · 3 stars · license MIT · pushed 2026-10-19 · branch master
API
GET https://markdownregistry.com/api/v1/artifacts/art_5qb5xx7dzakfaahx GET https://markdownregistry.com/api/v1/resolve?ref=aarontzeng/dev-lead/codex-adversarial-review GET https://markdownregistry.com/api/v1/blob/5139246ac78635bce9220885c223bed374b33305812f01978e540c8294f1d32b
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.