qa-debugging
Pass
Audited by Gen Agent Trust Hub on Apr 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions and documentation promote industry-standard debugging methodologies (scientific method, binary search, delta debugging) and modern observability practices.
- [SAFE]: Data safety is prioritized with explicit instructions across multiple files to redact personally identifiable information (PII), secrets, and credentials from logs and telemetry data.
- [SAFE]: The skill correctly identifies and warns the agent about the risk of indirect prompt injection when processing untrusted logs or artifacts.
- [SAFE]: All external URLs and dependencies referenced in the skill originate from trusted organizations (e.g., Google, Microsoft, OWASP) or well-known, reputable technology services (e.g., OpenTelemetry, SignOz, Cursor).
- [SAFE]: Shell commands and code snippets provided are illustrative examples for common debugging tasks and do not contain malicious execution or data exfiltration patterns.
Audit Metadata