bio-reporting-automated-qc-reports

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWPROMPT_INJECTION
Full Analysis
  • [Prompt Injection] (LOW): The skill has an indirect prompt injection surface as it processes external bioinformatics data to generate reports.
  • Ingestion points: The tool reads data from the results/ directory and user-provided custom TSV files.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded content are provided in the skill prompt.
  • Capability inventory: The skill executes the multiqc CLI and Python API, which involves reading multiple files and writing an aggregated HTML report to the filesystem.
  • Sanitization: The skill relies on the underlying multiqc library to parse and sanitize input data.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 08:03 AM