python-observability-ops
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill demonstrates legitimate observability patterns using standard libraries. It includes a specific redaction filter in
references/structured-logging.mddesigned to prevent sensitive information like passwords and API keys from being recorded in logs.\n- [INDIRECT_PROMPT_INJECTION]: The skill defines patterns for handling application logs and metrics which serve as ingestion points for external data.\n - Ingestion points:
references/structured-logging.md(log events),references/metrics.md(metric labels),references/tracing.md(span attributes).\n - Boundary markers: Uses structured logging and OpenTelemetry APIs which naturally separate metadata from message content.\n
- Capability inventory: The skill examples include network exports to observability collectors (e.g., OTLP/Prometheus endpoints).\n
- Sanitization: Proactively includes a
filter_sensitive_dataprocessor inreferences/structured-logging.mdthat uses regex and key-based filtering to redact sensitive information before it reaches the logging sink.
Audit Metadata