SkillSpector · gmail-poller

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

scripts/poller.py

Without declared permissions the skill's intent is opaque and cannot be validated.

SKILL.md

Model output is used without validation or sanitization. Unvalidated output injected into downstream contexts (SQL, shell, HTML) enables injection attacks and arbitrary code execution.

scripts/poller.py

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

credentials.yaml

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

credentials.yaml

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

credentials.yaml

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

dockerfile.fragment

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

dockerfile.fragment

Tool calls are chained to bypass individual safety checks or escalate capabilities beyond what any single tool call would allow.

dockerfile.fragment