add-agent
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONCREDENTIALS_UNSAFEDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
curlto perform HTTP requests (POST, GET, DELETE) against a LiteLLM proxy for agent lifecycle management. - [CREDENTIALS_UNSAFE]: The instructions guide the agent to request a
LITELLM_API_KEYfrom the user to populate the Authorization header for proxy requests. - [DATA_EXFILTRATION]: Administrative API keys and agent configurations are transmitted to the user-provided
LITELLM_BASE_URLto perform management tasks on the proxy service.
Audit Metadata