The Agent Skills Directory

[COMMAND_EXECUTION]: The skill interacts with the HealthEx service by executing a pre-installed binary located at $JARVIS_BIN_DIR/healthex. It uses subcommands to manage OAuth authentication, connection status, and data retrieval.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process unstructured data from external health records, including clinical notes and health summaries. This represents an attack surface where malicious instructions could be embedded in patient records to influence the agent's behavior.
Ingestion points: Sensitive medical data including clinical notes and summaries are retrieved through the mcp-call tool as described in SKILL.md.
Boundary markers: The instructions lack specific requirements for using delimiters or boundary markers when interpolating clinical record content into the agent's context.
Capability inventory: The skill possesses the capability to execute shell commands via the healthex binary and perform authenticated network operations to the HealthEx API.
Sanitization: There is no evidence of content filtering or sanitization for the retrieved clinical notes before they are processed by the agent.

healthex