kanban-orchestrator
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The orchestrator ingests untrusted user input to decompose requests into tasks. This content is used to populate the body field in kanban_create calls, which are then processed by specialist agents. 1. Ingestion points: User-supplied goals and multi-lane requests described in the Decomposition playbook section. 2. Boundary markers: The instructions do not specify the use of delimiters or safety warnings when interpolating user data into task descriptions. 3. Capability inventory: The skill utilizes task creation, linking, and completion tools, along with CLI execution capabilities for environment management. 4. Sanitization: No sanitization or validation of the user-provided task content is mentioned.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute CLI commands such as hermes profile list and hermes kanban reclaim using its terminal tool. These commands are intended for environment discovery and task lifecycle management within the Hermes platform.
Audit Metadata