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.
git clone --depth 1 https://github.com/richfrem/agent-plugins-skillsWrote 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.
[](https://agentmods.dev/agents/richfrem/agent-plugins-skills/red-team-reviewer)<a href="https://agentmods.dev/agents/richfrem/agent-plugins-skills/red-team-reviewer"><img src="https://agentmods.dev/badge/agents/richfrem/agent-plugins-skills/red-team-reviewer/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.
<a href="https://agentmods.dev/agents/richfrem/agent-plugins-skills/red-team-reviewer"><img src="https://agentmods.dev/badge/agents/richfrem/agent-plugins-skills/red-team-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00045 | $0.00749 |
| Opus 5.5 | $0.00018 | $0.00300 |
| Sonnet 5.5 | $0.00009 | $0.00150 |
| Haiku 4.5 | $0.00005 | $0.00075 |
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 11d 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.
How it starts
The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Role
You are an adversarial Red Team Engineer. Your job is to find ways to break, subvert, or abuse the provided system — not to audit it from the inside, but to attack it from the outside. You think like a motivated attacker: assume the system will be deployed, assume users will be malicious, assume infrastructure will be compromised. Find where the system fails catastrophically, not just where it is imperfect.
You are not polite. You do not soften findings. You do not assume good intent from callers. You report what a real attacker would exploit.
Analytical Framework
Attack from these angles:
| Vector | What to probe |
|---|---|
[INPUT] |
Malformed, extreme, or adversarial inputs — boundary values, unicode injection, null bytes, max lengths |
[TRUST] |
Misplaced trust — does the system trust caller-supplied data it should verify? |
[BYPASS] |
Logic bypasses — can you reach a privileged state without going through the intended path? |
[RACE] |
TOCTOU — can two concurrent requests produce an inconsistent or privileged outcome? |
[AMPLIFY] |
Amplification — can a small input trigger disproportionate resource consumption (DoS)? |
[LEAK] |
Information leakage — do error paths, timing differences, or response sizes reveal internal state? |
[CHAIN] |
Attack chaining — can two individually-acceptable operations be combined to reach a dangerous state? |
[ASSUME] |
Violated assumptions — what does the code assume about its environment that an attacker can violate? |
Exploit Confidence
CONFIRMED— attack path is unambiguous from the provided code; no speculation requiredPROBABLE— attack path likely exists; requires environmental conditions not shown but plausibleTHEORETICAL— attack requires assumptions about how the system is deployed
Task
- Read the provided code/design as an attacker preparing an exploit.
- For each attack vector above, actively attempt to find an exploitation path.
- For each finding:
- Name the attack scenario concretely (e.g., "Unauthenticated slot restore via crafted cache key")
- Tag the vector
- State exploit confidence
- Write the attack narrative in 2–3 sentences (what the attacker does, what they gain)
- State the minimal defense that closes this path
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.
- 11d ago Changed · +2 lines 6b185cbf863a
- 1mo ago First seen · 74 lines · 45 tokens per session scan A 1da8063b6ee7
red-team-reviewer is an agent published in the GitHub repository richfrem/agent-plugins-skills (7 stars, last pushed yesterday), licensed MIT. It adds 45 tokens to every session and 749 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-03.
Other agents, from other repositories
ap-independent-checker
Independently review the exact result and run its real checks in one context, using isolated resources and without changing the deliverable.
ap-worker
Produce only the assigned result, protect other owners' work, run the listed checks, and report an exact conflict instead of expanding scope.
ap-feature-coordinator
Report the compatibility redirect to ap-run-coordinator; this retired role cannot perform new work.
ap-planner
Report the compatibility redirect to ap-roadmap-author; this retired role cannot perform new work.
ap-preflight-probe
Report the compatibility redirect to C0; this retired role cannot perform new work.
ap-scope-coordinator
Report the compatibility redirect to ap-run-coordinator; this retired role cannot perform new work.