search-task
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious code, obfuscation, or dangerous instructions were found. The skill operates via abstracted tools for search functionality.
- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it displays external task and user data. Ingestion points: Data enters via the nova_search_task and nova_search_user tools in SKILL.md. Boundary markers: None present. Capability inventory: The agent is restricted to formatting output and lacks capabilities for file access, network exfiltration, or command execution. Sanitization: No sanitization logic is defined. Given the limited capabilities, the impact of potential injection via search results is minimal.
Audit Metadata