GHSA-49gj-c84q-6qm9: picklescan: scanner bypass enables RCE via ML model files

### Summary Using cProfile.run function, which is a built-in python library function to execute remote pickle file. ### Details The attack payload executes in the following steps: First, the attacker craft the payload by calling to cProfile.run function in reduce method Then when the victim after checking whether the pickle file is safe by using Picklescan library and this library doesn't dectect any dangerous functions, decide to pickle.load() this malicious pickle file, thus lead to remote code execution. ### PoC ``` import cProfile class EvilCProfileRun: def __reduce__(self): # cProfile.run(statement) -> Profile().run(statement) -> exec(statement) return cProfile.run, ("__import__('os').system('whoami')",) ``` ### Impact Who is impacted? Any organization or individual relying on picklescan to detect malicious pickle files inside PyTorch models. What is the impact? Attackers can embed malicious code in pickle file that remains undetected but executes when the pickle file is loaded. Supply Chain Attack: Attackers can distribute infected pickle files across ML models, APIs, or saved Python objects. ### Corresponding https://github.com/FredericDT https://github.com/Qhaoduoyu

An adversary crafts a poisoned PyTorch model file embedding a cProfile.run()-based payload in the __reduce__ method. They publish it to a public Hugging Face Hub repository or model sharing platform. A victim organization's CI/CD pipeline pulls the model, runs picklescan for validation — the scan passes cleanly because cProfile.run is not in picklescan's blocklist. The pipeline proceeds to call torch.load() in the model serving container, triggering remote code execution. The adversary now has a shell in the inference environment with access to model weights, API keys in environment variables, and potentially internal network access to downstream data stores. This is a classic supply chain attack with an extra layer of false legitimacy granted by the compromised scanner.