Skip to content

Unstick a queued agent

This guide frees an agent shown as QUEUED. It assumes the agent finished its dependencies and waits: dependency, bead, and time waits clear on their own and are not covered here.

Before you start

You need sase agent list (or the Agents tab) showing the stuck run. A row that ends with held by … waits on a hold; any other QUEUED row waits on runner capacity.

1. Clear a capacity wait

Check the fleet load in the Agents header (load: 8/10 means 8 of 10 capacity units are taken) and pick one:

  • Wait for a slot: waiters start in priority order as running agents finish, with no action needed.
  • Raise the cap briefly: press Ctrl+R in Launch Control to set a temporary max_running_agents. Parked agents react within about two seconds.
  • Kill an unwanted parked agent normally: dead waiter processes are ignored automatically and cannot wedge the queue.

Lowering the cap below current load is safe and never kills running work; admission simply pauses until capacity frees up. Confirm the agent leaves QUEUED on its next poll.

2. Clear a hold

A hold keeps its agents queued even when capacity is free. List active holds and release the one that matches:

sase agent hold list
sase agent hold release -k <armer-key>

The held agent is admitted on its next poll if capacity allows. Otherwise the hold ends by itself when its armer ends or its TTL expires. Report and prune holds whose armer died with:

sase doctor -C agent_holds.stale

If it goes wrong

A wait that matches neither capacity nor a hold is a different gate: run Diagnose problems with sase doctor and inspect the run's waiting.json as described in Runner slots.

_[project]: A sase project is a named unit of work registered with SASE. _[workspace]: A sase workspace is a numbered clone of a project's primary repo, managed by the workspace store and tracked in that project's registry.json. _[agent]: A sase agent is an agent session or a single agent that does not belong to a session. _[prompt]: A prompt is the text SASE hands to a provider to start an agent turn. _[macro]: A macro is a reusable prompt triggered with #name in agent prompts. _[Patch]: A Patch is SASE's local unit of change. _[bead]: A bead is one tracked work item in SASE's git-native issue tracker. _[plan]: A plan is a persisted, approved artifact recording the intent behind agent work. _[TUI]: The TUI is SASE's terminal interface for running agents and reviewing results. _[epic]: An epic is an executable multi-phase plan run with sase bead work. _[gate]: A sase gate is a durable, command-backed request for one user decision, created when work must pause for the user. _[monitor]: A sase monitor is a monitor turn that runs one long command under a detached supervisor, so the command outlives the agent that started it. *[stitch]: A stitch is the lightweight ordered change record inside a Patch's STITCHES: section.