using-agent-relay
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
agent-relayCLI and anadd_agenttool to manage and delegate tasks to workers.\n - The
add_agenttool executes provider-backed workers using a specified CLI and task description.\n - These operations are central to the skill's purpose of enabling agent orchestration and are consistent with the vendor's documentation.\n- [INDIRECT_PROMPT_INJECTION]: The skill facilitates reading messages from external sources, which creates a potential surface for indirect prompt injection.\n
- Ingestion points: Messages are retrieved via
check_inbox,list_messages, andget_message_thread(SKILL.md).\n - Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for processing message content.\n
- Capability inventory: The agent can perform actions via
add_agent,invoke_action, and messaging tools (SKILL.md).\n - Sanitization: No explicit content filtering or sanitization steps are documented for the ingested messages.
Audit Metadata