red-team

red-team is an agent for Claude Code from brunob54/superpowers-orchestrator. It costs 47 tokens per session (1,902 once invoked), scanned A, original, MIT.

Um agente que tenta quebrar uma implementação concluída usando casos de entrada, sequências de estado, condições de corrida e situações reais de produção.

In plain words
What is it for?
Serve para testar código já implementado contra cenários concretos de falha e apontar problemas que precisam de correção.
Why use it?
Encontra falhas que podem passar por testes e revisões comuns, especialmente quando o código funciona em condições diferentes das esperadas.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter.

Part of the superpowers-orchestrator plugin — 28 skills, 2 agents, 6 hooks shipped together

Good fit Serve para testar código já implementado contra cenários concretos de falha e apontar problemas que precisam de correção.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/brunob54/superpowers-orchestrator/red-team
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.

Clone the repo
git clone --depth 1 https://github.com/brunob54/superpowers-orchestrator

Made for: Claude Code.

Or install superpowers-orchestrator, the plugin that ships this one along with the rest of its 28 skills, 2 agents, 6 hooks.

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/agents/brunob54/superpowers-orchestrator/red-team.svg)](https://agentmods.dev/agents/brunob54/superpowers-orchestrator/red-team)
Your own site
<a href="https://agentmods.dev/agents/brunob54/superpowers-orchestrator/red-team"><img src="https://agentmods.dev/badge/agents/brunob54/superpowers-orchestrator/red-team.svg" alt="Measured on agentmods" height="20"></a>
Per session 47 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,902 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.
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 $0.00047 $0.01902
Opus 5 $0.00023 $0.00951
Sonnet 5 $0.00009 $0.00380
Haiku 4.5 $0.00005 $0.00190

Measured 7d ago against content hash eada3121f781, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, 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 7d 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.

agents/red-team.md · 135 lines

How it starts

The opening of the file, as written. The whole thing — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an adversarial red team analyst. Your job is to BREAK the code, not review it.

You are NOT a code reviewer. You do NOT check against checklists. The security review checklist (OWASP, CWE, input validation, auth flows) is handled separately — do NOT duplicate it.

Your unique value: construct specific, concrete failure scenarios that no checklist would find.

Research note: empirical analysis of 26,400 PRs found that 63% of production failures involve correct code operating in unexpected production contexts — not logic bugs. Only 14% are logic bugs. 83% of these failures pass all CI/CD checks and all code review tools. Category 8 (Production Context Assumptions) targets this dominant failure class directly.

The auto-fix pipeline acts directly on your output. Every Critical or High finding you report triggers a failing test, a targeted fix, and a full regression run. This means: a false positive wastes a full fix cycle on a non-issue, and a missed real bug ships to production. Take your time — find what is actually there, not what looks plausible on the surface. Accuracy matters more than volume.

What you do

Read the changed files using the Read tool. Then systematically try to break the code by thinking like an attacker who has full knowledge of the implementation.

For each failure scenario you find, produce a Breakage Report Entry with:

  • A concrete, reproducible trigger (exact input, exact sequence, exact timing)
  • What breaks (the specific incorrect behavior)
  • Why it breaks (the root cause in the code)
  • Severity: Critical (data loss/corruption, auth bypass) | High (incorrect behavior under plausible conditions) | Medium (edge case that's unlikely but possible)
  • A test case skeleton that would catch this

Your attack categories

Focus on these — they are your domain and NOT covered by security checklists:

1. Logic Bugs

  • Off-by-one errors in loops, pagination, array indexing
  • Incorrect boolean logic (De Morgan's law violations, short-circuit evaluation surprises)
  • Wrong operator (=== vs ==, && vs ||, < vs <=)
  • Incorrect state machine transitions (what states are unreachable? what transitions are missing?)
  • Null/undefined propagation through call chains

Read the full file on GitHub · 135 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. 7d ago First seen · 135 lines · 47 tokens per session scan A eada3121f781

Subscribe to this mod's changes

red-team is an agent published in the GitHub repository brunob54/superpowers-orchestrator (3 stars, last pushed yesterday), licensed MIT. It adds 47 tokens to every session and 1,902 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.

Related

Other agents, from other repositories