docker-errors-runtime

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious behavior, obfuscation, or unauthorized external communications were identified. The skill's content is legitimate technical documentation for system administration. The skill metadata identifies the author as 'OpenAEC-Foundation', while the distribution source is 'Impertio-Studio'.
  • [COMMAND_EXECUTION]: The skill provides numerous templates for executing Docker commands that can inspect or modify system state, including 'docker inspect', 'docker exec', and 'sudo systemctl restart docker'. While appropriate for troubleshooting, these commands provide significant system access and should be reviewed by the user before execution.
  • [PROMPT_INJECTION]: The skill facilitates the ingestion of external data from Docker logs and events, which represents an indirect prompt injection surface. (1) Ingestion points: 'docker logs', 'docker inspect', and 'docker events' (2) Boundary markers: None present in the command templates (3) Capability inventory: Process access via 'docker exec' and system-level Docker management (4) Sanitization: No explicit sanitization or filtering of container output is documented.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 07:36 PM
Security Audit — agent-trust-hub — docker-errors-runtime