personal-genomics-analysis
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches clinical variant data from the National Center for Biotechnology Information (NCBI) official FTP repository (ftp.ncbi.nlm.nih.gov). This is a well-known scientific service.
- [COMMAND_EXECUTION]: Requires the execution of local Python scripts to parse genomic data, manage SQLite databases, and generate summary reports.
- [PROMPT_INJECTION]: The skill ingests untrusted external data in the form of raw genetic data files, creating a surface for indirect prompt injection.
- Ingestion points: The parse_and_insert function in SKILL.md reads and processes a user-provided text file.
- Boundary markers: Absent; the skill does not use specific delimiters or instructions to prevent the agent from interpreting file content as executable instructions.
- Capability inventory: Local script execution, file system operations for database management, and network requests for downloading reference data.
- Sanitization: Implements basic structural validation of SNP identifiers and genotypes but lacks sanitization for potential embedded natural language instructions.
Audit Metadata