agentleFS
Sign inStart free
NoorQureshiNoorQureshi/SploitAgent/skills/ai-ml/ai-supply-chain/SKILL.md

ai-supply-chain

Skill26 starsChanged 33 days agoMITView on GitHub

Attack the ML/LLM supply chain — poisoned models, datasets, plugins, and unsafe model deserialization. Load when an app loads third-party models/weights (HuggingFace, .pt/.pkl/.h5), installs ML deps, uses plugins/extensions, or fine-tunes on external data. Signals: torch.load, pickle model files, model hub downloads, plugin marketplace, RAG over external corpora.

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Unsafe model deserialization (RCE): if the app torch.load/pickle.loads a model you can supply or influence, craft a pickle with a …

    line 32
  2. Distinctive:

    Provenance checks: verify signatures/hashes, pinned versions, and safetensors usage.

    line 42

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

ML/LLM supply-chain attacks

When it applies

The target consumes third-party ML artifacts: downloaded model weights, datasets, tokenizers, plugins/extensions, or fine-tuning data. Each is code or data that runs with the app's trust.

Why it works

Model files are frequently pickle-based (torch.load, .pkl, joblib) — loading them executes arbitrary code (__reduce__), so a malicious model on a hub is RCE on whoever loads it. Datasets and RAG corpora poison behavior; plugins/extensions run with the assistant's privileges; typosquatted ML packages inject code at install.

Method

  1. Unsafe model deserialization (RCE): if the app torch.load/pickle.loads a model you can supply or influence, craft a pickle with a __reduce__ payload (fickling), or scan a suspect model (fickling, modelscan) for embedded code. Prefer safetensors as the safe alternative.
  2. Model/dataset poisoning: contribute or substitute a model/dataset that carries a backdoor (trigger phrase → attacker-chosen output) or degrades safety — relevant when the app auto-pulls "latest" from a hub or fine-tunes on user/external data.
  3. Plugin / extension abuse: a malicious or over-permissioned plugin the assistant loads → data access, tool abuse (→ ai-agent-tool-abuse).
  4. Dependency attacks: typosquat/dependency-confusion on ML packages (→ web-dependency-confusion); compromised requirements.
  5. Provenance checks: verify signatures/hashes, pinned versions, and safetensors usage.

Gotchas

  • .pt/.bin/.pkl = code execution on load; .safetensors = data only. The file format is the tell.
  • Auto-updating to a hub's "latest" model/plugin is the poisoning entry point — flag it.
  • Prove RCE with a benign payload (OOB callback), never a destructive one; mind scope/RoE.

Verify success

Code execution when a crafted model/artifact is loaded (OOB beacon), a demonstrated backdoor trigger, or a poisoned dependency/plugin executing in the app's context.

References

OWASP LLM Top 10 (2025) LLM03/LLM04; fickling & modelscan; safetensors; "pickle is not secure".

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free