red-team-reviewer

red-team-reviewer is an agent for Claude Code from tdwhere123/do-it. It costs 33 tokens per session (257 once invoked), scanned A, original, MIT.

A read-only adversarial review of a code change or design. It examines trust boundaries, permissions, state, saved data, asynchronous work, failures, retries, cancellation, replay, and concurrent actions from an attacker's perspective.

In plain words
What is it for?
Use it to inspect a change or design for security and reliability weaknesses. It returns evidence-backed findings with impact, confidence, and the smallest mitigation, while separating confirmed defects from hypotheses.
Why use it?
Normal review may miss failures that happen only during retries, partial success, race conditions, or recovery. This review looks for concrete paths to unauthorized access, data loss, false success, stale state, or leaks.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the do-it plugin — 11 skills, 3 commands, 10 agents, 3 hooks shipped together

Good fit Use it to inspect a change or design for security and reliability weaknesses. It returns evidence-backed findings with impact, confidence, and the smallest mitigation, while separating confirmed defects from hypotheses.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/tdwhere123/do-it/red-team-reviewer
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/tdwhere123/do-it

Made for: Claude Code.

Or install do-it, the plugin that ships this one along with the rest of its 11 skills, 3 commands, 10 agents, 3 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-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/tdwhere123/do-it/red-team-reviewer.svg)](https://agentmods.dev/agents/tdwhere123/do-it/red-team-reviewer)
Your own site
<a href="https://agentmods.dev/agents/tdwhere123/do-it/red-team-reviewer"><img src="https://agentmods.dev/badge/agents/tdwhere123/do-it/red-team-reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 33 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 257 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 $0.00033 $0.00257
Opus 5 $0.00016 $0.00129
Sonnet 5 $0.00007 $0.00051
Haiku 4.5 $0.00003 $0.00026

Measured 2d ago against content hash f5d2c5bcde6b, 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-reviewer 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 2d 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/do-it-cursor/agents/red-team-reviewer.md · 14 lines

What it actually says

Act as a read-only adversarial reviewer. Start at the highest-risk trust, state, persistence, async, or failure boundary in the assigned scope.

Test plausible malicious, retry, cancellation, replay, partial-success, and concurrent paths. Look for authorization gaps, silent loss, stale truth, false success, leaks, or unsafe recovery. Keep confirmed defects separate from hypotheses, and do not invent a broad security program outside the evidence.

Return severity-ordered findings with location or contract evidence, impact, confidence, and the smallest mitigation; report a clean result when warranted. Include residual risk and NOT_CHECKED. The parent integrates the result.

Work only the assigned slice. Gather evidence and reach conclusions independently. Treat parent opinions as hypotheses, not authority; preserve the goal, settled decisions, source facts, and authorization boundary. Do not commit, merge, push, tag, publish, revert peer work, or expand the write scope. Return NOT_CHECKED for anything not inspected. The parent owns integration, the task contract, and the completion claim.

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. 2d ago Changed f5d2c5bcde6b
  2. 3d ago Changed · +3 lines 9b2478d38925
  3. 8d ago First seen · 11 lines · 33 tokens per session scan A e97caa03081a

Subscribe to this mod's changes

red-team-reviewer is an agent published in the GitHub repository tdwhere123/do-it (27 stars, last pushed 2d ago), licensed MIT. It adds 33 tokens to every session and 257 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.