doca-container-deployment
Pass
Audited by Gen Agent Trust Hub on Jul 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it requires the agent to ingest and interpret data from external sources that could be influenced by an attacker.\n
- Ingestion points: The agent is instructed to read YAML pod specifications and container log streams to verify deployment status and diagnose errors (TASKS.md, CAPABILITIES.md).\n
- Boundary markers: The skill contains explicit instructions to verify all configurations against official NVIDIA documentation and avoid 'inventing' parameters, which helps mitigate the risk of following injected instructions.\n
- Capability inventory: The skill utilizes shell commands to inspect files, check pod status, and retrieve logs.\n
- Sanitization: There is no evidence of programmatic sanitization for the log content being processed.
Audit Metadata