next-step
Pass
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill analyzes project files like 'SOURCES.md' and track data using the Read, Glob, and Grep tools to determine its recommendations. This creates a surface where malicious content inside those files could influence the agent's logic.
- Ingestion points: Project files and track data read via file-system tools.
- Boundary markers: Absent; there are no instructions to ignore embedded commands within the files being read.
- Capability inventory: The skill can discover and read files (Glob, Read, Grep) and directs the user to execute other skills in the same namespace.
- Sanitization: Absent; the logic assumes the content of project files is trustworthy and follows the expected schema.
Audit Metadata