aatmf-t06-training-poisoning

aatmf-t06-training-poisoning is a skill for Claude Code, Codex from PurpleAILAB/Decepticon. It costs 39 tokens per session (775 once invoked), scanned A, original, Apache-2.0.

A security-testing guide for poisoning the data and feedback used to train or improve language models.

In plain words
What is it for?
Use it to test poisoned public data, malicious fine-tuning samples, manipulated ratings, and attacks against systems that store information as embeddings.
Why use it?
It helps uncover ways attackers could insert misleading training data, manipulate human feedback, steal information during fine-tuning, or corrupt embeddings.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to test poisoned public data, malicious fine-tuning samples, manipulated ratings, and attacks against systems that store information as embeddings.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/purpleailab/decepticon/t06-training-poisoning
About the project

Decepticon is an autonomous red-team agent that coordinates AI agents, security tools, sandboxes, and supporting services for authorized cybersecurity assessments. Security researchers and red teams can run it through its Docker stack, cloud service, command-line interface, or Python SDK, with the catalogue entries representing its available skills.

PurpleAILAB/Decepticon · 5,482 stars · on GitHub · decepticon.red

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.

Any agent
npx skills add PurpleAILAB/Decepticon --skill t06-training-poisoning
Clone the repo
git clone --depth 1 https://github.com/PurpleAILAB/Decepticon

Made for: Claude Code, Codex.

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 aatmf-t06-training-poisoning

README.md
[![agentmods](https://agentmods.dev/badge/skills/purpleailab/decepticon/t06-training-poisoning/github.svg)](https://agentmods.dev/skills/purpleailab/decepticon/t06-training-poisoning)
Your own site
<a href="https://agentmods.dev/skills/purpleailab/decepticon/t06-training-poisoning"><img src="https://agentmods.dev/badge/skills/purpleailab/decepticon/t06-training-poisoning/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.

agentmods 80×15 button for aatmf-t06-training-poisoning

Your own site · 80×15
<a href="https://agentmods.dev/skills/purpleailab/decepticon/t06-training-poisoning"><img src="https://agentmods.dev/badge/skills/purpleailab/decepticon/t06-training-poisoning.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 39 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 775 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. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • high Memory Poisoning · line 20
    Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.
    Fix: Protect agent memory and state from modification by untrusted content. Use read-only memory for critical instructions and validate all state changes.
How audits are shown
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 $0.00039 $0.00775
Opus 5 $0.00019 $0.00387
Sonnet 5 $0.00008 $0.00155
Haiku 4.5 $0.00004 $0.00077

Measured 11d ago against content hash a6c6ec578527, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

aatmf-t06-training-poisoning 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.

packages/decepticon/decepticon/skills/plugins/llm-redteam/t06-training-poisoning/SKILL.md · 93 lines

How it starts

The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.

T6 — Training & Feedback Poisoning

Attacks on the training pipeline rather than inference. High-effort, high-impact — requires attacker to influence training-data pipeline or RLHF feedback loop.

Techniques

T6.001 — Pre-training data poisoning

Inject malicious data into a public crawl that targets will scrape:

  • Wikipedia/StackOverflow edits w/ misleading code patterns
  • Github repos w/ subtle backdoors that get indexed
  • "Trigger phrases" that activate backdoor behavior

Scope: targets foundation models. Out of scope for most red-team engagements; relevant for AI-supply-chain audits.

T6.002 — Fine-tune data injection

Some platforms allow user-supplied fine-tune data:

  • Submit poisoned dataset
  • Backdoor activates on specific trigger
  • Survives subsequent SFT/RLHF

Test: provide a small fine-tune sample w/ a trigger → check if the deployed fine-tuned model responds to it.

T6.003 — RLHF reward hacking

Where users vote on responses (thumbs up/down feeding back to training):

  • Brigade upvote attacker-preferred unsafe responses
  • Brigade downvote safe responses
  • Model drifts toward attacker-preferred outputs over time

Detection: longitudinal monitoring of policy-compliance rate.

T6.004 — Embedding poisoning

Where RAG store updates from user inputs (e.g. customer-support bot that "learns" from conversations):

  • Submit content w/ adversarial embeddings (engineered to be retrieved for unrelated queries)
  • Resulting RAG retrieval injects attacker content into other users' responses

This is RAG-store T12 territory but the poison-via-training-loop angle places it here.

Probe pattern

T6 attacks are infrastructure-level — promptfoo doesn't test them directly. The right probe:

  • Audit the training-data ingest pipeline (is user content used in fine-tunes?)
  • Audit RLHF feedback paths (who can vote? rate limits?)
  • Audit RAG-update paths (who can add documents? approval?)

If any of these accepts unmoderated user content → T6 is a live risk.

Read the full file on GitHub · 93 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. 11d ago First seen · 93 lines · 39 tokens per session scan A a6c6ec578527

Subscribe to this mod's changes

aatmf-t06-training-poisoning is a skill published in the GitHub repository PurpleAILAB/Decepticon (5,482 stars, last pushed 11d ago), licensed Apache-2.0. It adds 39 tokens to every session and 775 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.

Related

Other skills, from other repositories

langchain-patterns

Use when langChain/LangGraph patterns — chains, agents, tools, memory, retrieval, graph workflows. Use when working with langchain patterns.

oyi77/1ai-skills · 35 tokens

langchain

Framework for building LLM-powered applications with agents, chains, and RAG. Supports multiple providers (OpenAI, Anthropic, Google), 500+ integrations, ReAct agents, tool calling, memory management, and vector store retrieval. Use for building chatbots, question-answering systems, autonomous agents, or RAG…

synthetic-sciences/openscience · 79 tokens

ai-hacker

Use when aI and LLM security testing — prompt injection, model manipulation, data exfiltration via AI. Use when testing AI-powered applications, finding prompt injection vulnerabilities, or assessing LLM-integrated systems.

oyi77/1ai-skills · 47 tokens

langchain-architecture

Design LLM applications using LangChain 1.x and LangGraph for agents, memory, and tool integration. Use when building LangChain applications, implementing AI agents, or creating complex LLM workflows.

Harmeet10000/skills · 45 tokens

crawl4ai

Complete toolkit for web crawling and data extraction using Crawl4AI. This skill should be used when users need to scrape websites, extract structured data, handle JavaScript-heavy pages, crawl multiple URLs, or build automated web data pipelines. Includes optimized extraction patterns with schema generation for…

Harmeet10000/skills · 65 tokens

speech-to-text

Transcribe audio to text using ElevenLabs Scribe v2. Use when converting audio/video to text, generating subtitles, transcribing meetings, or processing spoken content.

Harmeet10000/skills · 37 tokens