The Agent Skills Directory

[SAFE]: The skill is a standard implementation of clinical guidelines and does not contain any malicious patterns.
[DATA_EXFILTRATION]: No network operations (e.g., curl, requests) or access to sensitive local files (e.g., ~/.ssh, .env) were detected. The script only processes provided checklist data to generate status reports.
[COMMAND_EXECUTION]: The skill does not perform any system-level command execution. There are no uses of os.system, subprocess, or other shell invocation methods.
[OBFUSCATION]: No Base64-encoded strings, zero-width characters, or homoglyph-based URL obfuscation were found in any of the skill files.
[INDIRECT_PROMPT_INJECTION]: The skill's ingestion of external data is limited to specific checklist fields with no high-risk capabilities.
Ingestion points: Data enters via patient_name, procedure, and responses arguments in the script.py and skill.py entry points.
Boundary markers: The instructions clearly define phase-specific contexts (Sign In, Time Out, Sign Out) and timing.
Capability inventory: The skill has no capabilities for file writing, network transmission, or code execution.
Sanitization: The script processes input data into structured JSON objects using standard library functions.

surgical-safety-checklist