After every task ticket linked to the spec has been closed following its merged PR, should the orchestrator close the parent spec ticket? Decide what to retain or clean up: .orchestrate/<spec-ticket-number> files/branch, orchestration worktrees, task worktrees, Herdr panes, and logs. Recovery must remain possible until the run is complete, including from another workstation.
Part of #27
## Question
After every task ticket linked to the spec has been closed following its merged PR, should the orchestrator close the parent spec ticket? Decide what to retain or clean up: `.orchestrate/<spec-ticket-number>` files/branch, orchestration worktrees, task worktrees, Herdr panes, and logs. Recovery must remain possible until the run is complete, including from another workstation.
Close the parent spec ticket only after every in-scope task is complete, its PR is merged, required checks/reviews pass, and run state is safely persisted. Keep it open while work is blocked, failed, or unresolved.
Use these terms: orchestrator system = where the skill is invoked; worker node = the user-selected machine that runs subagents and repository work (local or remote); worker agent = a subagent running on that node.
The worker node is authoritative for .orchestrate/<slug>. A new orchestrator system reconnects to that node and syncs a local copy from it. Keep handoffs and logs with the state; no separate Git branch is needed just to transfer state.
At completion, remove clean task worktrees and close only Herdr panes created by this run. Preserve dirty or unresolved worktrees. Retain the worker node's canonical checkout and orchestration state.
Never delete retained state automatically; cleanup requires an explicit user action.
Decision:
- Close the parent spec ticket only after every in-scope task is complete, its PR is merged, required checks/reviews pass, and run state is safely persisted. Keep it open while work is blocked, failed, or unresolved.
- Use these terms: orchestrator system = where the skill is invoked; worker node = the user-selected machine that runs subagents and repository work (local or remote); worker agent = a subagent running on that node.
- The worker node is authoritative for `.orchestrate/<slug>`. A new orchestrator system reconnects to that node and syncs a local copy from it. Keep handoffs and logs with the state; no separate Git branch is needed just to transfer state.
- At completion, remove clean task worktrees and close only Herdr panes created by this run. Preserve dirty or unresolved worktrees. Retain the worker node's canonical checkout and orchestration state.
- Never delete retained state automatically; cleanup requires an explicit user action.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Part of #27
Question
After every task ticket linked to the spec has been closed following its merged PR, should the orchestrator close the parent spec ticket? Decide what to retain or clean up:
.orchestrate/<spec-ticket-number>files/branch, orchestration worktrees, task worktrees, Herdr panes, and logs. Recovery must remain possible until the run is complete, including from another workstation.Decision:
.orchestrate/<slug>. A new orchestrator system reconnects to that node and syncs a local copy from it. Keep handoffs and logs with the state; no separate Git branch is needed just to transfer state.