observability

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [SAFE]: The skill documents best practices for observability and does not contain malicious code, obfuscation, or unauthorized data access patterns.
  • [INDIRECT_PROMPT_INJECTION]: The skill documents patterns for building observability systems that ingest external data. Ingestion points: Middleware captures HTTP request and response data (node-patterns.md). Boundary markers: The skill advises using allowlists for named fields (SKILL.md). Capability inventory: Suggestions include using the OTel SDK and logging infrastructure (node-patterns.md). Sanitization: The skill mandates redacting PII at the source and provides testing patterns for hygiene (SKILL.md, testing-telemetry.md).
  • [EXTERNAL_DOWNLOADS]: All external references are to well-known and trusted technology documentation and blogs, including Google SRE, Stripe, and OpenTelemetry official sites.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 05:44 PM
Security Audit — agent-trust-hub — observability