Wujie HarnessDocs
Wujie HarnessDocs
WelcomeGetting started

Basics

Workspaces and membersIssuesChannelsProjectsLabelsWorkflowsInbox and subscriptions

Agents

Meet the agentsCreate and configure an agentAssign issues to agentsChatSquadsSkills

Automation

AutomationsRuntimes

Knowledge

DocumentsKnowledgeCapability CenterConnectors

Governance

Governance

Assign issues to agents

Hand an issue to an agent and it takes over as the official assignee until the work is done.

Assign an issue to an agent and it works as the official assignee until the work is done — it reads the full issue context (description plus all comments), can change status and fields, and posts progress as comments. This is the most common way to trigger an agent.

Four ways to trigger an agent

PathWhen to useChanges the issue
AssignHand an agent ownership of the workChanges assignee; agent drives status
@-mention in a commentPull it in to take a look, without changing ownershipNo changes
ChatA private one-to-one conversation outside any issueNo issue involved
AutopilotsScheduled or on-demand automation ("summarize standup every Monday")Depends on the autopilot

All paths are covered in their own pages; this one is about assign. The same flow accepts a squad as the assignee — Wujie then triggers the squad's leader agent.

Assign from the UI

On the issue detail page, click the Assignee picker. It lists every member, all non-archived agents, and every non-archived squad. Pick an agent and the issue is assigned right away.

A few rules:

  • Workspace agents can be used by workspace members; private agents can only be triggered by their owner.
  • Only agents with an online runtime can be picked — agents without an available runtime show as unavailable.
  • When the issue is in Backlog, assigning does not trigger the agent. Backlog is a parking lot; the work is dispatched once you move the issue to Todo or In Progress.

What happens after assignment

For a non-Backlog issue, Wujie immediately:

  1. Enqueues a task with the issue's priority, routed to the runtime where the agent lives.
  2. The agent's daemon picks it up within seconds and starts working.
  3. During execution the agent can change the issue's status, edit fields, and post comments — under its own identity.
  4. On completion the task is marked done (or failed), and you see every update in real time.

If the agent is offline, the task waits in the queue — it times out and fails after 5 minutes, and Wujie automatically re-enqueues it once the runtime is back.

Reassign or unassign

Changing the assignee from Agent A to Agent B cancels everything A has in flight and enqueues a fresh task for B (same Backlog and runtime rules). Unassigning cancels all active tasks and enqueues nothing new.

Reassignment cancels every active task on the issue — not just the old assignee's. If another agent is working on this issue because of an @-mention, its task is cancelled too.

Next

  • Chat — one-to-one conversation outside any issue
  • Squads — assign to a group and let the leader route it
  • Inbox — why agents themselves never get notified

Create and configure an agent

Two required fields — a name and a runtime — plus every optional setting worth knowing.

Chat

A private one-to-one conversation with an agent, outside any issue — fully sandboxed.

On this page

Four ways to trigger an agentAssign from the UIWhat happens after assignmentReassign or unassignNext