The Agent Skills Directory

[COMMAND_EXECUTION] (SAFE): The scripts check_completeness.py and generate_template.py perform basic file I/O and regex operations. While generate_template.py allows defining an output path, its behavior is consistent with a template generator and lacks any mechanism for arbitrary shell command execution.
[DATA_EXPOSURE & EXFILTRATION] (SAFE): No hardcoded credentials, sensitive system paths, or network communication patterns were identified. The scripts exclusively process user-provided LaTeX files or internal assets.
[INDIRECT_PROMPT_INJECTION] (SAFE): The check_completeness.py script identifies and displays uncustomized placeholders from input files. This represents a minor data ingestion surface, but the script contains no logic that would interpret or execute the content of those files. 1. Ingestion points: scripts/check_completeness.py reads user-provided files for validation. 2. Boundary markers: Absent. 3. Capability inventory: Limited to file reading and printing results to the console. 4. Sanitization: The script uses regex to filter for specific patterns, effectively limiting the scope of ingested data to expected structural elements.

treatment-plans