The Agent Skills Directory

[COMMAND_EXECUTION] (HIGH): The skill requests the Bash tool and provides instructions to execute a script (python scripts/security_scan.py <project_path>). This pattern allows the agent to run code against arbitrary user-controlled directories, which is high-risk if the script behavior is not strictly constrained.
[PROMPT_INJECTION] (HIGH): This skill is vulnerable to Indirect Prompt Injection (Category 8) because it processes untrusted data with high-privilege capabilities.
Ingestion points: The agent uses Read, Glob, and Grep to scan files within a user's <project_path>.
Boundary markers: No specific boundary markers or instructions to ignore embedded commands within the scanned files are present in the skill definition.
Capability inventory: The agent has access to the Bash tool and the ability to execute Python scripts.
Sanitization: There is no evidence of sanitization or filtering of the content read from the project files before the agent processes it.
[REMOTE_CODE_EXECUTION] (MEDIUM): While no direct remote downloads are seen in the markdown, the skill references an external script (scripts/security_scan.py) that is not provided for analysis. If this script is dynamically updated or retrieved from an untrusted source, it constitutes a Category 4/10 risk.

vulnerability-scanner