signals-scout-error-tracking

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from error reports, which could be leveraged to perform indirect prompt injection attacks.
  • Ingestion points: The agent reads external and potentially attacker-controlled data from stack traces, issue descriptions, and activity logs via tools like query-error-tracking-issue, execute-sql (against the events table), and advanced-activity-logs-list.
  • Boundary markers: The instructions do not define clear delimiters or provide the agent with specific 'ignore' warnings for instructions that might be embedded within the error data or logs it processes.
  • Capability inventory: The skill has significant capabilities including writing to the report channel (emit_report, edit_report) and persisting data across sessions via the scratchpad (scout-scratchpad-remember).
  • Sanitization: There is no mention of sanitizing or filtering the content extracted from error stack traces or logs before the agent processes or reports on it.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 01:03 AM
Security Audit — agent-trust-hub — signals-scout-error-tracking