delete-team
Pass
Audited by Gen Agent Trust Hub on Apr 6, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted data from an external API which creates an indirect prompt injection surface.
- Ingestion points: Team metadata is retrieved from the /team/list endpoint in SKILL.md.
- Boundary markers: No delimiters or protective instructions are used to isolate the API response from the agent's prompt instructions.
- Capability inventory: The agent uses Bash and curl to perform destructive POST requests to delete teams.
- Sanitization: No sanitization is performed on the team aliases or IDs before the agent processes them for confirmation.
- [COMMAND_EXECUTION]: The skill utilizes curl to perform administrative operations on a LiteLLM proxy, which is consistent with its stated purpose by the vendor.
Audit Metadata