grilling
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions explicitly direct the agent to gather facts from the environment, including the filesystem and tools, to answer questions. This creates an attack surface where the agent could ingest data containing malicious instructions from the files it processes.
- Ingestion points: Filesystem and tool outputs as specified in the instruction to "dispatch a sub-agent to find [facts]" in
SKILL.md. - Boundary markers: The instructions do not define any delimiters or warnings for the agent to ignore instructions embedded within the data it finds.
- Capability inventory: The skill encourages the use of file-reading and tool-execution capabilities to autonomously gather information.
- Sanitization: There is no mention of sanitizing or validating the information retrieved from the environment before the agent uses it to formulate new questions.
Audit Metadata