signals-scout-csp-violations

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted event data from $csp_violation events to generate security reports. Ingestion points: Properties such as $csp_blocked_url, $csp_document_url, and $csp_source_file in the events table (queried via execute-sql) which are generated by external user browsers. Boundary markers: The instructions do not define delimiters or specific instructions for the agent to ignore potentially malicious commands embedded in the ingested event data. Capability inventory: The skill has the ability to query the analytics database (execute-sql) and create or modify entries in the platform's report inbox (emit_report, edit_report). Sanitization: There is no evidence of explicit sanitization, escaping, or filtering of externally sourced event properties before they are interpolated into the agent's context for analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 09:40 AM
Security Audit — agent-trust-hub — signals-scout-csp-violations