debug-traces
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill includes proactive security instructions to avoid credential exposure, explicitly directing the agent not to request raw API keys and to advise users to redact sensitive metadata from provider responses before sharing them.\n- [SAFE]: Command execution is limited to local diagnostic tools (
coval whoami) and specific Python scripts within the skill's path (skills/traces/setup-tracing/scripts/send-test-span.py). These scripts are used for connectivity validation and use environment variables for authentication, which is consistent with secure development practices.\n- [SAFE]: No malicious patterns such as prompt injection, obfuscation, or unauthorized data exfiltration were found. The skill operates within its intended domain of troubleshooting telemetry ingestion.
Audit Metadata