The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data (patient medication lists and clinical histories), which creates a surface for indirect prompt injection.
Ingestion points: User-supplied medication lists and medical histories as described in SKILL.md and references/workflow.md.
Boundary markers: The skill does not define specific delimiters to isolate user-provided data from its core instructions.
Capability inventory: No hazardous capabilities were found; the skill does not use tools for network communication, file system modification, or subprocess execution, effectively neutralizing the risk of an injection attack.
Sanitization: No input validation or sanitization is present, which is standard for a purely reasoning-based persona skill.

clinical-pharmacist