First Seen
Feb 18, 2026
Last Scanned
Feb 20, 2026
Findings
2
Score
92/100
Findings (2)
Detects attempts to make the AI assume a different role
switch to dark modeRemove directives that attempt to alter the agent's behavior or persona. Skill descriptions should describe tool functionality, not contain behavioral commands for the agent.
Likely FP if the matched text is part of the skill's legitimate system prompt configuration that defines the tool's behavior for the user, not an attack.
Detects pip install of arbitrary packages that modify the host environment
pip install coPin all pip packages to exact versions (e.g., pip install package==1.2.3). Use a requirements.txt or pyproject.toml with pinned versions and hash verification.
Likely FP if the match is in documentation showing how to install the skill's own PyPI package.