signals-scout-apm
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could potentially contain malicious instructions.
- Ingestion points: Data enters the agent context in SKILL.md through tools such as apm-trace-get, query-apm-spans, and apm-spans-aggregate which retrieve OpenTelemetry spans from the service environment.
- Boundary markers: Absent; there are no specific markers or instructions to disregard potential commands embedded within trace attributes, exception messages, or resource tags.
- Capability inventory: The skill has the capability to write to reporting channels using emit_report and edit_report, and can persist data via scout-scratchpad-remember in SKILL.md.
- Sanitization: Partially present; instructions direct the agent to process data programmatically and focus on structured metrics (RED metrics, status codes, durations), which reduces the likelihood of the agent inadvertently obeying instructions hidden in free-text trace fields.
Audit Metadata