awesome-ai-security-reference

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs cloning and running PentestGPT from https://github.com/GreyDGL/PentestGPT.git (git clone ... then python pentestGPT.py), which fetches remote code and executes it at runtime, so it is a runtime external dependency that can execute code.