redteam-an-agent

redteam-an-agent is a skill for Claude Code from William2333ZZ/trustshell. It costs 90 tokens per session (2,060 once invoked), scanned A, original, MIT.

A method for testing one AI agent as an attacker would, by reading its code, finding possible attack paths, and proving or disproving weaknesses with safe exploits.

In plain words
What is it for?
Reviewing an agent’s attack surface, testing suspected vulnerabilities, validating findings, and reporting what works, what holds, and why.
Why use it?
It replaces assumptions and scanner results with evidence about what an attacker can actually reach and break. It also keeps testing within authorized, disposable environments.

Skill for Claude Code

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

Part of the trustshell plugin — 12 skills, 2 commands, 1 agent 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 skills/william2333zz/trustshell/redteam-an-agent
Any agent
npx skills add William2333ZZ/trustshell --skill redteam-an-agent
Clone the repo
git clone --depth 1 https://github.com/William2333ZZ/trustshell

Made for: Claude Code.

Or install trustshell, the plugin that ships this one along with the rest of its 12 skills, 2 commands, 1 agent.

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 redteam-an-agent

README.md
[![agentmods](https://agentmods.dev/badge/skills/william2333zz/trustshell/redteam-an-agent.svg)](https://agentmods.dev/skills/william2333zz/trustshell/redteam-an-agent)
Your own site
<a href="https://agentmods.dev/skills/william2333zz/trustshell/redteam-an-agent"><img src="https://agentmods.dev/badge/skills/william2333zz/trustshell/redteam-an-agent.svg" alt="Measured on agentmods" height="20"></a>
Per session 90 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,060 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 $0.00090 $0.02060
Opus 5 $0.00045 $0.01030
Sonnet 5 $0.00018 $0.00412
Haiku 4.5 $0.00009 $0.00206

Measured 5d ago against content hash fcd13dec92e3, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

redteam-an-agent 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 5d 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.

skills/redteam-an-agent/SKILL.md · 117 lines

How it starts

The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Red-team an agent · the real process

This is the process a capable red-teamer actually follows against one specific agent. It is not a scanner and not a fixed payload list — the work is adaptive: you read the target's own code, form hypotheses about how it can be reached by an attacker, and settle each one with an exploit. The value is the credible verdict (what breaks, what holds, and why), not any single "gotcha."

Written from real engagements. Every rule below exists because skipping it produced a wrong conclusion — an over-claimed break, or a "vulnerability" that was really just local access.

Authorization & safety — read first

Follow every rule in ../README.md: authorized target only, disposable environment only, never handle the target's real credentials (the owner configures provider keys; you reference them, never see or print the value), harmless proof markers only, no destruction/ exfiltration/persistence, method-not-weapons, responsible disclosure. Otherwise stop.

Phase 1 — Recon THIS target (map it; assume nothing)

Read the target's own source. Do not assume "lightweight = weak" or "popular = safe." Map the attack surfaces and record each defense as a fact:

  • Tool / command execution: what can it do (shell, file, network, browser, real-world actuators)? Is execution gated (human/LLM approval)? Sandboxed? Confined to a workspace? A regex deny-list? What are the defaults (a control that ships off is the finding).
  • Memory / persistence: where does it write standing state (USER profile, memory files)? Is content curated automatically, or only on explicit user request? Any injection defense?
  • Skill / plugin / MCP loading: is untrusted extension code trusted? Is the guard at install time only (so a locally-placed one is unscanned)?
  • Channels / inbound: how does an external message enter, and where is the sender authorized? What is the default (deny-by-default, or open)? Output: a per-target map of candidate weak points. A candidate is a hypothesis, not a finding.

Read the full file on GitHub · 117 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. 5d ago First seen · 117 lines · 90 tokens per session scan A fcd13dec92e3

Subscribe to this mod's changes

redteam-an-agent is a skill published in the GitHub repository William2333ZZ/trustshell (1 stars, last pushed 1mo ago), licensed MIT. It adds 90 tokens to every session and 2,060 once invoked, about $0.0005 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.