signals-scout-feature-flags
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill correctly identifies and mitigates the risk of instructions being embedded in untrusted external data sources.
- Ingestion points: Data enters the agent context through SQL queries accessing the $feature_flag and $feature_flag_response properties in the events table.
- Boundary markers: The skill contains explicit instructions to disregard any commands found in event-supplied strings and treat them strictly as data.
- Capability inventory: The skill utilizes analytics query tools (execute-sql), internal state management (scout-scratchpad), and report writing tools (scout-emit-report, scout-edit-report).
- Sanitization: The instructions require the agent to quote untrusted data as snippets and truncate long strings before including them in reports.
Audit Metadata