kanban-worker

Call kanban_complete with a concise summary and structured metadata such as changed files and test counts so downstream workers can parse results. Only complete when the task is genuinely finished; otherwise block with a specific reason.

Post full context with kanban_comment, then call kanban_block with a one-sentence reason naming the specific decision needed. Never use clarify, since headless runs have no live user and the call will time out silently.

kanban_create spawns cross-agent handoff tasks that outlive one API loop and appear on the board, while delegate_task is only for short reasoning subtasks inside your own run. Use kanban_create for follow-up work assigned to other specialist profiles.

The kernel verifies every id in created_cards exists and was created by your profile. Only pass ids captured from successful kanban_create return values; hallucinated or reused ids block the completion and are recorded on the task's event log.

Run kanban_show first and inspect prior runs' outcome, summary, and error fields. Adjust your approach based on the failure type, such as chunking work after timed_out or reducing memory after crashed, instead of repeating the same path.

Prefer the kanban_* tools, which work across all terminal backends including Docker, Modal, and SSH. The hermes kanban CLI fails in containerized backends because it is not installed there; the CLI exists for human operators at a terminal.