The Agent Skills Directory

[SAFE]: The skill is primarily instructional and focused on generating documentation and architecture plans for healthcare systems. It does not contain any executable scripts, network requests, or sensitive file access.
[PROMPT_INJECTION]: No patterns of instruction overriding, jailbreaking, or safety filter bypasses were detected in the instructions or metadata.
[DATA_EXPOSURE_AND_EXFILTRATION]: No hardcoded credentials or sensitive file paths (like .ssh or .aws) are present. The skill explicitly instructs the agent to avoid generating real Protected Health Information (PHI) and to use mock data for testing purposes.
[COMMAND_EXECUTION]: The skill does not use any shell commands, subprocess calls, or system-level operations.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests architectural requirements from the user, it possesses no dangerous capabilities (such as code execution, file writes, or external API calls) that could be exploited through malicious data input. It uses a structured template (FHIR_INTEGRATION_TEMPLATE.md) which helps delineate between instructions and data.

healthcare-fhir