add-team
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for administrative management of a LiteLLM proxy. It follows best practices by requesting the 'LITELLM_BASE_URL' and 'LITELLM_API_KEY' from the user at runtime rather than using hardcoded credentials or attempting to access sensitive local configuration files.
- [SAFE]: The skill configuration uses 'allowed-tools' to restrict the execution environment to only the 'curl' command within the Bash tool, adhering to the principle of least privilege.
- [DATA_EXPOSURE_AND_EXFILTRATION]: While the skill handles an API key, it does so to authenticate requests to a destination URL provided and controlled by the user. The communication is limited to the user's own proxy infrastructure.
- [INDIRECT_PROMPT_INJECTION]: The skill interpolates user-provided data (team name, budget) into a JSON payload. While this creates a theoretical surface for injection if the input contains escaped characters, the impact is confined to the specific API request initiated by the user for their own proxy management.
Audit Metadata