dsh-goal-quiescence
Run the following command in DeepSeek Harness:
dsh plugin install 1052326311/dsh-goal-quiescence
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install 1052326311/dsh-goal-quiescence in DeepSeek Harness to install; the source repository is at https://github.com/1052326311/dsh-goal-quiescence.
About this plugin
In multi-agent workflows, a parent agent may call update_goal(action=complete) right after delegating background reviews, only to discover that child agents have not yet finished. Their results then arrive too late to influence the already-closed goal. dsh-goal-quiescence plugs exactly that "settled but unseen" race window.
The plugin enforces a lifecycle gate: it records the start/end pair of every subagent run that begins after it loads, then denies completion until each run has settled and the parent has explicitly called goal_quiescence_ack to pull back the terminal output into its own context. goal_quiescence_status lists whichever runs still block completion. It neither schedules, cancels, nor retries children, and it does not replace Harness goal mode; it simply guards the single "complete" transition.
Ideal for developers who routinely delegate background reviews, code audits, or parallel subtasks under Harness goal mode and have experienced the awkward situation of a goal marked done while a child report is still in flight.
Use Cases
- Prevent a goal from closing before delegated background reviews return.
- Block premature completion when parallel subtasks have settled but their output is unreviewed.
- Ensure every child agent terminal output lands in the parent context before the goal is finalized.
Best For
- Developers who frequently delegate background subtasks under Harness goal mode.
- Workflow orchestrators who need auditable, traceable multi-agent pipelines.
- Engineering teams that have hit the goal closed but child report unread problem.
Related Plugins
A method pack that makes AI coding agents plan against your real baseline, prove completion with fresh evidence, and reduce reworks and unsafe changes.
Turns the DeepSeek Harness session into a captain that builds a durable sub-agent team, splits goals into dependency-aware tasks, and coordinates work via direct messages and a live Web UI.
Gives coding agents design judgment, letting Claude Code, Cursor, and 70+ agents generate and iterate high-quality UI, presentations, and graphics right from the terminal.
Run the Pi ecosystem's plugins on DeepSeek Harness, unmodified, via a compatibility layer that implements Pi's public extension ABI on DSH's native services.