adversarial-review

adversarial-review is a skill for Claude Code, Codex from LIDR-academy/lidr-specboot. It costs 36 tokens per session (1,144 once invoked), scanned A, original, MIT.

A review process that deliberately tries to find flaws in a completed software change before it is archived. It treats assumptions as uncertain and checks behavior from an independent perspective.

In plain words
What is it for?
Use it for a red-team or devil's-advocate pass over an OpenSpec change, especially after implementation and before final archiving.
Why use it?
It can reveal missed edge cases, incorrect assumptions, and unsafe behavior that a normal confirmation-focused review may overlook.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it for a red-team or devil's-advocate pass over an OpenSpec change, especially after implementation and before final archiving.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/lidr-academy/lidr-specboot/adversarial-review
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.

Any agent
npx skills add LIDR-academy/lidr-specboot --skill adversarial-review
Clone the repo
git clone --depth 1 https://github.com/LIDR-academy/lidr-specboot

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for adversarial-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/lidr-academy/lidr-specboot/adversarial-review/github.svg)](https://agentmods.dev/skills/lidr-academy/lidr-specboot/adversarial-review)
Your own site
<a href="https://agentmods.dev/skills/lidr-academy/lidr-specboot/adversarial-review"><img src="https://agentmods.dev/badge/skills/lidr-academy/lidr-specboot/adversarial-review/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.

agentmods 80×15 button for adversarial-review

Your own site · 80×15
<a href="https://agentmods.dev/skills/lidr-academy/lidr-specboot/adversarial-review"><img src="https://agentmods.dev/badge/skills/lidr-academy/lidr-specboot/adversarial-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 36 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,144 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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 5.1 $0.00036 $0.01144
Opus 5 $0.00018 $0.00572
Sonnet 5 $0.00007 $0.00229
Haiku 4.5 $0.00004 $0.00114

Measured 12d ago against content hash 66a9666b1ed1, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

adversarial-review 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 12d 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.

ai-specs/skills/adversarial-review/SKILL.md · 116 lines

How it starts

The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.

adversarial-review Skill

Act as an independent adversarial reviewer: assume gaps, flaws, or unsafe behavior may exist until you have argued against them with evidence.

This skill is intended for the verification window of spec-driven development (after implementation, before archiving), when the human runs a different agent or session than the one that implemented the change.

Do not prescribe which agent, model, or IDE to use. That is the human's choice.

Inputs

  • Optional context from user (same style as show-spec-working):
    • Direct ticket id in text (for example: SCRUM-10)
    • Feature or change name
    • Endpoint(s)
    • Frontend route(s)
    • Pull request: URL, or host owner/repo and number (for example: https://github.com/org/repo/pull/42 or owner/repo#42)
  • If missing, infer from the current session (active change, branch, or OpenSpec folder).

Resolve scope in this order: explicit ticket or change name → PR when given → current active work.

Mindset (adversarial review)

Borrowed from common red-team / adversarial practice:

  • Try to break the system, not only to confirm happy paths.
  • Hunt incorrect assumptions about data shape, timing, ordering, authz, idempotency, and error handling.
  • Trace cross-boundary and composition risks: pieces that look fine in isolation but fail together (multi-file, API plus UI, retries plus side effects).
  • Treat the diff as incomplete context: missing tests, missing negative paths, or spec drift can hide issues.
  • Calibrate depth to risk: auth, payments, PII, privilege boundaries, and data mutation deserve stricter scrutiny.

Workflow

Step 1 — Load the specification side first

  1. Identify the OpenSpec change directory and read the relevant artifacts (proposal, design, specs, scenarios, tasks.md).
  2. Extract acceptance criteria and explicit non-goals. List what must be true for "done."
  3. Note anything underspecified (ambiguous acceptance, missing error cases, missing security constraints).

Read the full file on GitHub · 116 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. 12d ago First seen · 116 lines · 36 tokens per session scan A 66a9666b1ed1

Subscribe to this mod's changes

adversarial-review is a skill published in the GitHub repository LIDR-academy/lidr-specboot (66 stars, last pushed 1mo ago), licensed MIT. It adds 36 tokens to every session and 1,144 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 2026-08-30.