The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it ingests untrusted clinical data and user-provided diagram descriptions which are then used to influence agent behavior.
Ingestion points: Clinical reports, case study data, and natural language diagram descriptions provided to the schematic generation tool.
Boundary markers: No explicit boundary markers or 'ignore' instructions are defined for the clinical data being processed.
Capability inventory: The skill uses the Bash tool to execute Python scripts and the Write/Edit tools to modify files.
Sanitization: There is no evidence of input sanitization for the natural language strings passed to the shell-based schematic generator.
[COMMAND_EXECUTION]: The skill documentation explicitly instructs the agent to use the Bash tool to execute local scripts (e.g., python scripts/generate_schematic.py "your diagram description"). This creates a command execution surface where malicious natural language input could potentially be used for command injection if the underlying script does not properly handle shell escaping.

clinical-reports