ask-nico
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data by reading the instruction files (SKILL.md) of other installed skills and project-level configuration to provide recommendations.
- Ingestion points: Step 2 of the skill instructions (SKILL.md) requires the agent to read discoverable installed skill entries and project authority pointers from the environment.
- Boundary markers: The instructions do not define technical delimiters for ingested content but explicitly instruct the agent to verify claims by reading full source files rather than trusting summary descriptions.
- Capability inventory: The agent is strictly limited by instructions to an advisory and read-only role, with clear prohibitions against invoking tools, updating state, or making network calls.
- Sanitization: No specific content filtering or sanitization mechanisms are defined for the data read from external skills.
Audit Metadata