managed-model-endpoints
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides legitimate guidelines for managing model endpoints and infrastructure without introducing malicious code or configurations.
- [SAFE]: The skill promotes safe data handling by explicitly instructing the agent to store secret references as environment variables or credential references rather than in plain-text metadata, which follows standard security practices.
- [PROMPT_INJECTION]: The workflow includes a step to capture model inference responses for verification, which creates a potential surface for indirect prompt injection. Ingestion points: Inference response data as described in SKILL.md workflow step 5. Boundary markers: No explicit delimiters for external content are specified. Capability inventory: Description mentions startup scripts and inference routing. Sanitization: No specific filtering or sanitization of the captured responses is mentioned. This risk is assessed as safe given the limited scope of the auditing task.
Audit Metadata