The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script at ~/ai_projects/claude-skills/health-data/scripts/health_query.py to retrieve vitals, workouts, sleep, and activity data. It also passes custom SQL queries as arguments to this script to calculate specific health metrics like average blood pressure.
[DATA_EXFILTRATION]: The skill accesses sensitive personal health information, including heart rate, blood pressure, and sleep records from a local database. While this is the primary intended purpose for personalization, it involves reading sensitive health data files.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes content from external web sources and local health records, creating a potential injection surface.
Ingestion points: External research papers and expert analysis from Tier 1-4 domains (PubMed, Mayo Clinic, etc.) and health data records retrieved via script execution.
Boundary markers: Absent. The instructions do not specify the use of delimiters or explicit instructions to ignore embedded commands within the content retrieved from external search tools.
Capability inventory: Includes web searching (WebSearch, Tavily), full-text research (Firecrawl), and shell command execution (Python script) for local data access.
Sanitization: No sanitization or filtering of external content is identified before the synthesis of the final evidence-based response.

doctorg