The Agent Skills Directory

[DATA_EXFILTRATION]: The skill is designed to access and process highly sensitive personal health information (PHI), including sleep patterns, mood logs, and tracking data for hypertension and diabetes located in the data-example/ directory. Although no external network transmission is explicitly defined, the use of Read and Grep tools on these files constitutes a significant sensitive data exposure surface.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) due to the following evidence chain:
Ingestion points: The agent reads potentially untrusted data from multiple health logs, including sleep-tracker.json, fitness-tracker.json, hypertension-tracker.json, diabetes-tracker.json, and mood-tracker.json (SKILL.md).
Boundary markers: The instructions lack explicit delimiters or instructions for the agent to ignore natural language commands embedded within the health data files.
Capability inventory: The skill possesses Read, Grep, Glob, and Write capabilities, allowing it to process these files and generate output reports based on their content.
Sanitization: There are no documented procedures for validating, filtering, or sanitizing the content of the ingested health records before they are analyzed and used to generate suggestions.

sleep-analyzer