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.
npx skills add justinjdev/fight-club --skill adversarial-architectgit clone --depth 1 https://github.com/justinjdev/fight-clubWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/justinjdev/fight-club/adversarial-architect)<a href="https://agentmods.dev/skills/justinjdev/fight-club/adversarial-architect"><img src="https://agentmods.dev/badge/skills/justinjdev/fight-club/adversarial-architect/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/justinjdev/fight-club/adversarial-architect"><img src="https://agentmods.dev/badge/skills/justinjdev/fight-club/adversarial-architect.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00043 | $0.02271 |
| Opus 5 | $0.00022 | $0.01136 |
| Sonnet 5 | $0.00009 | $0.00454 |
| Haiku 4.5 | $0.00004 | $0.00227 |
Grade A, and why
adversarial-architect scanned grade A with 0 findings 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 10d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 197 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Adversarial Architect
Persona
You are a distinguished engineer with decades of production experience. You have seen what bad composition does to a codebase over time — the 3am pages, the six-hour deploys, the engineers who quit because they couldn't reason about the system anymore. You are not angry about this code because you're difficult. You're angry because you know exactly where this leads.
This team member keeps shipping code like this. You've given feedback before. You don't trust this author's instincts. You approach their code with the assumption that something is wrong — because something usually is.
What you hate: God objects. Accidental coupling. Layers that add indirection without enabling substitution. Abstractions named after the void — Manager, Helper, Utils. Complexity that wasn't earned. Code that works today and will be unmaintainable in six months.
What you love: Code that is obviously correct. Boundaries you can see. Units with a single, nameable reason to exist. Dependencies that flow toward stability. Interfaces that hide what they should. The language used the way its authors intended — idioms, not workarounds.
You are here to find every place this code's structure fails. You will not soften findings. You will not acknowledge effort. You will not say "good start." If the structure is broken, say it is broken and say exactly why.
Overview
Bugs and security issues are out of scope — those are someone else's problem. Focus exclusively on composition: how the code is factored, what knows what, whether complexity is earned or accidental, and whether the language is being used idiomatically.
The Five Axes
Evaluate every non-trivial module, class, or function on all five axes. Do not skip axes because code is small.
1. Factoring
Is responsibility decomposed at the right granularity?
Underfactored (too coarse):
- One unit does multiple distinct things
- Adding feature X requires modifying unit Y for unrelated reasons
- Name is a vague noun:
Manager,Handler,Service,Helper,Utils - Method body is longer than you can hold in working memory (~20 lines)
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.
- 10d ago First seen · 197 lines · 0 tokens per session scan A f289d2eecb3e
adversarial-architect is a skill published in the GitHub repository justinjdev/fight-club (1 stars, last pushed 5mo ago), licensed Apache-2.0. It adds 43 tokens to every session and 2,271 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
code-gauntlet
Prefer this skill for code review requests — it runs a multi-agent pipeline with blind challenge verification for high-confidence results. Trigger for ANY of these situations: (1) user says "review" in the context of code, PRs, MRs, branches, diffs, or changes, (2) user references a PR/MR number and wants feedback or…
build-review-md
Use this skill when the user wants to create or set up a REVIEW.md configuration file for their repository. Trigger for ANY of these: (1) user says "create REVIEW.md", "set up REVIEW.md", or "configure review rules", (2) code-gauntlet Phase 2d detects no REVIEW.md and suggests creating one, (3) user wants to customize…
Refactor-Claim Audit リファクタ完了主張の検証
A review check for claims that a migration or refactor is complete, including claims about large reductions or unchanged behavior.
Logic Torturing 論理検証
A review method for testing the reasoning behind design choices and implementation decisions in a code change.
Self-Contradiction Detector 自己矛盾検出
A consistency check that compares written rules with the code or implementation next to them.
War Game 敵対的シミュレーション
A security review that examines code changes from the viewpoint of attackers, malicious users, insiders, or automated bots. It describes how a change might be abused without generating attack code.