manual-instrumentation
Pass
Audited by Gen Agent Trust Hub on Apr 23, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No override instructions, bypass markers, or safety guideline violations were detected. The instructions focus on guiding the agent through instrumentation design.
- [DATA_EXFILTRATION]: The skill mentions configuring OTLP exporters for telemetry data, which is standard behavior for OpenTelemetry. No hardcoded malicious endpoints or instructions to exfiltrate sensitive files were found.
- [REMOTE_CODE_EXECUTION]: No commands for downloading and executing remote scripts (e.g., curl|bash) or installing external packages from untrusted sources were found.
- [OBFUSCATION]: The content is clear and uses standard Markdown. No Base64, hex encoding, zero-width characters, or hidden text patterns were detected.
- [INDIRECT_PROMPT_INJECTION]: While the skill processes user code for instrumentation reviews, it does not possess capabilities that would lead to high-risk escalation. It functions primarily as a guidance and verification tool.
Audit Metadata