observability-engineer
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external telemetry data (logs, metrics, traces), which presents a surface for indirect prompt injection if that data contains malicious instructions.
- Ingestion points: Processing of external logs, metrics, and distributed traces from various platforms (SKILL.md).
- Boundary markers: Absent. There are no explicit instructions to the agent to disregard instructions embedded within the ingested telemetry data.
- Capability inventory: None. The skill provides no scripts or tools to execute these tasks directly.
- Sanitization: Absent. While it mentions log parsing and structuring, it does not specify sanitization against instruction injection.
- [NO_CODE]: The skill consists entirely of Markdown instructions and persona descriptions. No scripts, binaries, or configuration files are provided for execution.
Audit Metadata