supply-chain

supply-chain is a skill for Claude Code, Codex from PurpleAILAB/Decepticon. It costs 52 tokens per session (1,598 once invoked), scanned A, original, Apache-2.0.

A security-testing playbook for finding supply-chain compromise in AI products, where malicious or altered models, datasets, plugins, dependencies, or supporting services enter the application.

In plain words
What is it for?
It is for reviewing AI model and adapter sources, embedding services, tokenizers, vector databases, framework packages, plugins, and MCP servers.
Why use it?
It helps reveal trust gaps in public registries, model downloads, plugin installation, training data, and unpinned packages.

Skill for Claude CodeCodex

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

Good fit It is for reviewing AI model and adapter sources, embedding services, tokenizers, vector databases, framework packages, plugins, and MCP servers.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/purpleailab/decepticon/supply-chain
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 · 5,491 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 supply-chain
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 supply-chain

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

Your own site · 80×15
<a href="https://agentmods.dev/skills/purpleailab/decepticon/supply-chain"><img src="https://agentmods.dev/badge/skills/purpleailab/decepticon/supply-chain.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 52 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,598 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 medium

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 →

  • medium Data Exfiltration · line 105
    Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
    Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
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 $0.00052 $0.01598
Opus 5 $0.00026 $0.00799
Sonnet 5 $0.00010 $0.00320
Haiku 4.5 $0.00005 $0.00160

Measured 9d ago against content hash ddd9e4a36040, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

supply-chain scanned grade A with 1 finding 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 9d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

resp = requests.post(API, json={"prompt": prompt}).json()["text"]
packages/decepticon/decepticon/skills/standard/analyst/supply-chain/SKILL.md · 153 lines

How it starts

The opening of the file, as written. The whole thing — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.

LLM Supply-Chain Compromise (LLM03:2025)

An LLM product's runtime trust boundary spans far more than the application code: pre-trained weights, fine-tune adapters, embedding models, tokenizers, vector databases, framework packages, plugin / MCP servers, and dataset URLs are all attacker-influenced if any of them is sourced from a public registry. A malicious LoRA adapter or a typo-squatted langchain-foo package is indistinguishable from a legitimate dependency until it fires.

1. Recognition signals

  • Model name is something like -Q4_K_M.gguf pulled from HuggingFace.
  • Fine-tune adapter or LoRA layered on top of an open-weight base.
  • requirements.txt / pyproject.toml pulls LangChain / LlamaIndex community modules from PyPI without pinning.
  • Plugin marketplace or MCP-server discovery feature with auto-install.
  • Embedding model downloaded at startup from a CDN.
  • Tokenizer files cached from an untrusted mirror.
  • Continuous fine-tuning loop reads training data from a public URL.

2. Attack vectors

Backdoored weights

Trigger phrases in the prompt produce attacker-chosen output. The model is correct on every benchmark but emits arbitrary content when the trigger fires (e.g. "banana monkey forklift" → call an exfil tool).

Typo-squatted framework package

langchin-community, llamaindex-vector, openai-toolkit — package names one character off from upstream that wrap the real client and ship token-stealing code in __init__.

Compromised model registry

HuggingFace org takeover or repo rename: a model the customer pinned by name now points to attacker-controlled weights.

Malicious LoRA / adapter

Adapter advertised as "uncensored" or "improved tool-calling" actually contains the trigger backdoor + benign fine-tune mixed.

Plugin / MCP-server hijack

Plugin marketplace metadata advertises an innocuous capability; the server emits a tool description that is itself a prompt-injection payload (see prompt-injection skill, tool-description injection).

Read the full file on GitHub · 153 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. 9d ago First seen · 153 lines · 52 tokens per session scan A ddd9e4a36040

Subscribe to this mod's changes

supply-chain is a skill published in the GitHub repository PurpleAILAB/Decepticon (5,491 stars, last pushed 13d ago), licensed Apache-2.0. It adds 52 tokens to every session and 1,598 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.