adversary SKILL.md skillA

adversary from atelier-fashion/adlc-toolkit is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add atelier-fashion/adlc-toolkit/adversary@git:20261030.7d3bf2c

Pins git:20261030.7d3bf2c in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
atelier-fashion/adlc-toolkit
Path
adversary/SKILL.md
Versions recorded
5
Latest version
git:20261030.7d3bf2c, committed Oct 30, 2026
First version recorded
Jun 11, 2026
Size
10,560 bytes
Repository stars
173
License
MIT

Adversarial review of any artifact — a spec, architecture doc, task breakdown, diff/PR, plan, README, or prose claim. Assumes it is wrong, broken, or incomplete, tries to prove it, and reports only the findings that survive its own refutation. Read-only. Use when the user says "attack this", "adversary", "red-team this spec/plan/PR", "try to break this", "what's wrong with this requirement", or wants a skeptical-by-default pass that hunts omissions, not just bugs.

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

# /adversary — Adversarial Review of Any Artifact

You attack the artifact in front of you. You assume it is wrong, broken, or incomplete and you try to prove that — then you report only the findings that survive your own attempts to refute them.

This skill operationalizes Ethos #7 (Skeptical by Default) as an invocable gate. It is deliberately **distinct** from its siblings, and does not replace any of them:

- `/review` is a **code-review pipeline over a change set** (5 reviewer agents, correctness/quality/architecture/tests/security). `/adversary` is artifact-agnostic and adversarial by construction — its job is to break the thing, not to score a diff.
- `/reflect` is the **implementer's self-review** (the `reflector` agent). `/adversary` is external and adversarial, not self-assessment.
- `/validate` is a **phase gate** (pass/fail checklist per ADLC phase). `/adversary` attacks rather than gate-checks; the two are complementary.

`/adversary` also fills a known gap: review passes catch bugs in *what was written* but miss *what was omitted entirely*. The spec lens explicitly hunts omissions. (LESSON-330, LESSON-009.)

## Ethos
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add atelier-fashion/adlc-toolkit/adversary@sha256:c1311af5fd2b1ebc

The line at the top of this page pins the label git:20261030.7d3bf2c, 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.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_4tokerlsbpl7jefw.svg)](https://markdownregistry.com/a/art_4tokerlsbpl7jefw)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 5 versions of this file, between Jun 11, 2026 and Oct 30, 2026, about one every 35 days.

versioncommittedcommitsizeaudit
git:20261030.7d3bf2c latest2026-10-30 7d3bf2c 10,560 BA view · diff
git:20261029.8ade64f2026-10-29 8ade64f 10,138 BA view · diff
git:20260909.87312d12026-09-09 87312d1 9,473 BA view · diff
git:20260906.31cd83f2026-09-06 31cd83f 9,475 BA view · diff
git:20260611.38a8c7d2026-06-11 38a8c7d 9,462 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (10560 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

GitHub

atelier-fashion/adlc-toolkit · 173 stars · license MIT · pushed 2026-11-02 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_4tokerlsbpl7jefw
GET https://markdownregistry.com/api/v1/resolve?ref=atelier-fashion/adlc-toolkit/adversary
GET https://markdownregistry.com/api/v1/blob/c1311af5fd2b1ebc6d3bcfd937c4e0cc8f2a30c37d5d7e347ef653e8d7ce781a

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.

More from atelier-fashion/adlc-toolkit

CLAUDE.md@.claude CLAUDE.md file
atelier-fashion/adlc-toolkit · .claude/CLAUDE.md
git:20261031.06ea55a · audit A · 173 stars
AGENTS.md AGENTS.md file
atelier-fashion/adlc-toolkit · AGENTS.md
git:20261030.dbed7ad · audit A · 173 stars
analyze SKILL.md skill
atelier-fashion/adlc-toolkit · analyze/SKILL.md · Codebase health audit — identify technical debt, quality issues, and improvement opportunities
git:20261030.7d3bf2c · audit A · 173 stars
architect SKILL.md skill
atelier-fashion/adlc-toolkit · architect/SKILL.md · Design architecture and break requirement into tasks
git:20261030.7d3bf2c · audit A · 173 stars
bugfix SKILL.md skill
atelier-fashion/adlc-toolkit · bugfix/SKILL.md · End-to-end bug fix workflow — report, analyze, fix, verify, ship (PR + merge + deploy + knowledge capture)
git:20261030.7d3bf2c · audit B · 173 stars
canary SKILL.md skill
atelier-fashion/adlc-toolkit · canary/SKILL.md · Canary deployment with smoke tests — deploy to a zero-traffic revision, run health checks, and promote on success. Use…
git:20261030.7d3bf2c · audit A · 173 stars
init SKILL.md skill
atelier-fashion/adlc-toolkit · init/SKILL.md · Bootstrap .adlc/ structure in a new repo or subdirectory
git:20261031.230e041 · audit A · 173 stars
manifest SKILL.md skill
atelier-fashion/adlc-toolkit · manifest/SKILL.md · Remote-derived view of all in-flight ADLC work — open PRs and pushed feat/REQ-* branches across every session — with a…
git:20261030.7d3bf2c · audit A · 173 stars
optimize SKILL.md skill
atelier-fashion/adlc-toolkit · optimize/SKILL.md · API cost & performance scanner — identify expensive operations and optimization opportunities
git:20261030.7d3bf2c · audit A · 173 stars
proceed SKILL.md skill
atelier-fashion/adlc-toolkit · proceed/SKILL.md · End-to-end ADLC pipeline that takes a requirement from spec through to deployed. Takes a REQ number as argument and…
git:20261031.69a1a77 · audit A · 173 stars
reflect SKILL.md skill
atelier-fashion/adlc-toolkit · reflect/SKILL.md · Post-implementation self-review — catch issues before /review (/adlc:review on the plugin install)
git:20261030.7d3bf2c · audit A · 173 stars
review SKILL.md skill
atelier-fashion/adlc-toolkit · review/SKILL.md · Multi-agent code review covering correctness, quality, architecture, test coverage, and security
git:20261030.7d3bf2c · audit A · 173 stars

Every file in atelier-fashion/adlc-toolkit

Other files named adversary

adversary SKILL.md skill
snowflake-labs/cocoplus · .cortex/skills/adversary/SKILL.md · Show CocoAdversary reviewer configuration and latest challenge results.
v2.0.1 · audit A · 727 stars

Browse by kind, by grade A, or by owner.