sniff-report
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes and presents data derived from scanning external project files. If a scanned file contains malicious instructions (e.g., hidden in code comments or strings), these could potentially influence the agent's behavior during the reporting phase.
- Ingestion points: Scan results retrieved via the
sniffMCP tool or legacysniff_reportcommand. - Boundary markers: The instructions do not define specific delimiters or warnings to separate untrusted scan data from agent instructions during output generation.
- Capability inventory: The skill is restricted to reporting and does not perform file writes, network operations, or arbitrary command execution.
- Sanitization: No specific sanitization or filtering logic is mentioned for the retrieved report content.
Audit Metadata