core-kanban
Pass
Audited by Gen Agent Trust Hub on Jun 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a local shell script (../../scripts/lab-kanban.sh) for state management. This is the intended mechanism for the task board protocol and represents standard operational behavior.\n- [PROMPT_INJECTION]: The skill processes external research task data which acts as an indirect prompt injection surface. The agent may ingest content from these tasks. Evidence chain: (1) Ingestion points: Research task board file; (2) Boundary markers: Not specified in instructions; (3) Capability inventory: Executes local shell scripts via lab-kanban.sh; (4) Sanitization: None specified.
Audit Metadata