red-team

red-team is a command for Claude Code from aaronbassett/agent-foundry. It costs 37 tokens per session (898 once invoked), scanned A, original, MIT.

A review command that asks an adversarial software agent to look for the strongest reasons a proposed plan, answer, design, or other work could fail. It returns one of three decisions: accept, revise, or reject.

In plain words
What is it for?
Use it to challenge a plan, answer, design, or artifact when you want a final critical review before moving ahead.
Why use it?
It exposes risks and failure cases before you commit to a decision or implementation.

Command for Claude Code

Written for Claude Code: argument-hint in frontmatter. Also seen: mentions subagents.

Part of the decision-making plugin — 1 skill, 7 commands, 2 agents shipped together

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 commands/aaronbassett/agent-foundry/red-team
Clone the repo
git clone --depth 1 https://github.com/aaronbassett/agent-foundry

Made for: Claude Code.

Or install decision-making, the plugin that ships this one along with the rest of its 1 skill, 7 commands, 2 agents.

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 red-team

README.md
[![agentmods](https://agentmods.dev/badge/commands/aaronbassett/agent-foundry/red-team.svg)](https://agentmods.dev/commands/aaronbassett/agent-foundry/red-team)
Your own site
<a href="https://agentmods.dev/commands/aaronbassett/agent-foundry/red-team"><img src="https://agentmods.dev/badge/commands/aaronbassett/agent-foundry/red-team.svg" alt="Measured on agentmods" height="20"></a>
Per session 37 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 898 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 5.1 $0.00037 $0.00898
Opus 5 $0.00018 $0.00449
Sonnet 5 $0.00007 $0.00180
Haiku 4.5 $0.00004 $0.00090

Measured 6d ago against content hash 351936cfb14c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

red-team 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 6d 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.

plugins/decision-making/commands/red-team.md · 95 lines

How it starts

The opening of the file, as written. The whole thing — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/decision-making:red-team

When to use

Use this command when you have a proposed plan, answer, design, or artifact and want to stress-test it before committing. It is intended for the moment just before a decision is finalized — when the shape of the plan is clear enough to critique but not yet locked in.

Use /decision-making:pre-mortem instead for decisions that have already been tentatively made and you want to imagine how they failed in hindsight. red-team is for catching failure modes before commit; pre-mortem is for surfacing them after tentative commit.

Cost tier

Medium. One committed adversarial subagent, one round. See references/cost-tiers.md.

Input

The plan, answer, design, or artifact to be red-teamed, plus any relevant context the reviewer will need (constraints, assumptions, prior discussion, known trade-offs). If the caller's arguments do not include an artifact to review, ask once for it.

Workflow

  1. Spawn one devil's-advocate subagent — use the Agent tool with subagent_type: general-purpose. Pass the artifact and context into the bracketed slots of this prompt template verbatim:

    You are a devil's advocate reviewing a proposed [plan / answer /
    design / artifact]. Your sole mandate: find the strongest reasons
    this will fail.
    
    The artifact under review:
    [ARTIFACT]
    
    Context:
    [CONTEXT]
    
    Look for:
    - Hidden assumptions that may not hold
    - Edge cases the artifact does not account for
    - Second-order effects the author may not have considered
    - What a hostile reviewer would say in a code review
    - Failure modes under unusual load, adversarial input, or stress
    
    Return a ranked list of weaknesses. For each weakness, assign:
    - Severity (low | medium | high) — how bad would it be if it happened
    - Plausibility (low | medium | high) — how likely is it to happen
    
    Good-faith rules: no fabricated facts, no straw-manning, focus on
    real weaknesses (not nit-picks), commit fully to the adversarial
    stance without equivocation.
    

Read the full file on GitHub · 95 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. 6d ago First seen · 95 lines · 37 tokens per session scan A 351936cfb14c

Subscribe to this mod's changes

red-team is a command published in the GitHub repository aaronbassett/agent-foundry (4 stars, last pushed 21d ago), licensed MIT. It adds 37 tokens to every session and 898 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.