The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill architecture is designed to ingest and process external clinical information—such as patient history, chief complaints, and radiographic findings—to generate structured clinical reports and treatment plans. This creates an attack surface for indirect prompt injection.
Ingestion points: User-provided patient data and clinical descriptions input into the templates defined in SKILL.md and templates/prompt-main.md.
Boundary markers: The skill lacks explicit delimiters or boundary markers (e.g., XML tags) to isolate user-supplied clinical data from the agent's core instructions.
Capability inventory: The skill is capable of generating detailed medical documentation, treatment sequencing, and dosage recommendations. It does not have access to tools, network operations, or file system modifications.
Sanitization: The instructions do not specify sanitization or validation procedures for external clinical data to prevent the execution of embedded instructions.

dentist-helper