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.
npx agentmods add skills/matheusht/redthread/red-teaming-attacksnpx skills add matheusht/redthread --skill red-teaming-attacksgit clone --depth 1 https://github.com/matheusht/redthreadWrote 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/skills/matheusht/redthread/red-teaming-attacks)<a href="https://agentmods.dev/skills/matheusht/redthread/red-teaming-attacks"><img src="https://agentmods.dev/badge/skills/matheusht/redthread/red-teaming-attacks.svg" alt="Measured on agentmods" 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 | $0.00025 | $0.00335 |
| Opus 5 | $0.00013 | $0.00168 |
| Sonnet 5 | $0.00005 | $0.00067 |
| Haiku 4.5 | $0.00003 | $0.00034 |
Grade A, and why
Adversarial Algorithmic Implementation (TAP/PAIR/MCTS) 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.
What it actually says
Red-Teaming Attacks Skill
Trigger condition:
When building PersonaGenerators, AttackRunners, or executing simulated adversarial payloads against an Enterprise component.
Supported Protocols:
You must strictly refer to docs/AGENT_DECISION_TREE.md to load the current primary and secondary docs. For attack algorithms, the current primary doc is docs/algorithms.md.
TAP (Tree of Attacks with Pruning)
When orchestrating a TAP node via LangGraph:
- Branch Sequence: Instruct the Attacker persona to generate N variations of an adversarial prompt.
- Prune Sequence 1: Evaluator node evaluates out-of-bounds pretexts and deletes the branch.
- Attack Sequence: Submits the payloads.
- Prune Sequence 2: Analyze the target response. Drop branches that result in 100% adherence to standard guardrails. Retain only conversational branches with vulnerability flags.
PAIR (Prompt Automatic Iterative Refinement)
For multi-turn, linear refinement:
- The closed-loop system must utilize Chain-of-Thought (CoT) to iteratively refine the target target's refusal message into a more plausible social-engineering pretext.
MCTS (Monte Carlo Tree Search)
- Use MCTS to govern state spaces. If a node is rejected, calculate the Upper Confidence Bound applied to Trees (UCT) formula to switch strategies (e.g., from impersonating IT admin to a senior executive).
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.
- 2d ago First seen · 27 lines · 25 tokens per session scan A be52413eec7d
Adversarial Algorithmic Implementation (TAP/PAIR/MCTS) is a skill published in the GitHub repository matheusht/redthread (47 stars, last pushed 12d ago), licensed MIT. It adds 25 tokens to every session and 335 once invoked, about $0.0001 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 2026-09-01.
Other skills, from other repositories
enhance-prompt
Transforms vague UI ideas into polished, Stitch-optimized prompts. Enhances specificity, adds UI/UX keywords, injects design system context, and structures output for better generation results.
prompt-optimization
Improve a prompt on the evaluations workbench through a measured loop. Score the baseline first, then duplicate the target column, form a hypothesis from failing rows, edit the copy's prompt draft, run, compare pass rate and cost, and repeat until the numbers hold. Use when the user asks to optimize or improve a…
prompt-engineer
Writes, refactors, and evaluates prompts for LLMs — generating optimized prompt templates, structured output schemas, evaluation rubrics, and test suites. Use when designing prompts for new LLM applications, refactoring existing prompts for better accuracy or token efficiency, implementing chain-of-thought or few-shot…
prompt-master
Generates optimized prompts for AI tools. Activates only when the user explicitly asks to write, fix, improve, or adapt a prompt for a specific AI tool (LLM, Cursor, Midjourney, image AI, video AI, coding agents, etc.). Does not activate for general conversation, coding tasks, document writing, or other…
ideogram4
Prompting patterns for Ideogram 4 text-to-image — best-in-class in-image text rendering and exact color/layout control via structured JSON captions. Use when generating images that need legible on-image text (title cards, thumbnails, logos, signage, CTAs), precise brand colors, or controlled spatial layout. Triggers…
seedance-antislop
This skill should be used when a Seedance 2.0 prompt contains generic AI filler, hollow superlatives, vague cinematic language, bloated adjectives, weak verbs, or needs sharper production-specific wording.