stern-logs
Pass
Audited by Gen Agent Trust Hub on Aug 6, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill defines how the AI agent should invoke the 'stern' command. It mandates the use of '--no-follow' to ensure the agent does not hang on infinite log streams, which is a critical safety practice for automated contexts.
- [EXTERNAL_DOWNLOADS]: The documentation provides installation paths for 'stern' using trusted services like Homebrew, GitHub, and the GitHub Container Registry (GHCR).
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it processes untrusted log data from Kubernetes pods.
- Ingestion points: Application logs from Kubernetes containers at runtime.
- Boundary markers: The skill recommends using Go templates and specific flags like '--only-log-lines' to isolate log messages from metadata.
- Capability inventory: The agent has the ability to execute 'stern', 'kubectl', and log processors like 'jq'.
- Sanitization: The instructions explicitly advise the agent to be cautious about sensitive data (secrets) in logs and to quote content selectively.
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were identified. The skill follows best practices for cluster interaction and security awareness.
Audit Metadata