challenge
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze potentially untrusted data, including user arguments ($ARGUMENTS) and files identified through the Read, Glob, and Grep tools, creating a surface for indirect prompt injection attacks.
- Ingestion points: The skill processes a user-specified target in SKILL.md and allows a sub-agent to read relevant file paths in the 'deep' subcommand.
- Boundary markers: The instructions do not specify the use of delimiters (like XML tags or triple quotes) or 'ignore embedded instructions' warnings for the target content being analyzed.
- Capability inventory: The skill utilizes file system tools (Read, Glob, Grep), sub-agent spawning (Agent tool), and user interaction (AskUserQuestion).
- Sanitization: There are no explicit instructions for the agent to sanitize, escape, or filter instructions that might be embedded within the target text or files.
Audit Metadata