grilling
Pass
Audited by Gen Agent Trust Hub on Aug 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface for indirect prompt injection by instructing the agent to combine user-provided plans with data retrieved from the local environment.\n
- Ingestion points: User input describing plans or decisions and facts retrieved from the filesystem or tools by sub-agents (SKILL.md).\n
- Boundary markers: Absent. The skill does not define delimiters or instructions to ignore embedded commands for the data it retrieves from the environment (SKILL.md).\n
- Capability inventory: The agent is explicitly instructed to use sub-agents to access the filesystem and system tools to resolve frontier questions (SKILL.md).\n
- Sanitization: Absent. No mention of escaping, filtering, or validating content retrieved from external files or user input before processing (SKILL.md).
Audit Metadata