Agents and workers

AgentGrid distinguishes a master (or orchestrator) from the workers it starts. A master plans and coordinates; each worker handles a bounded task in its own visible pane.

Master and worker

• A master is a harness pane with AgentGrid's orchestration MCP enabled. There is no separate master pane type: a compatible agent pane becomes a master when it starts or associates another pane.
• A worker is an agent the master starts with spawn_worker, spawn_role, or resume_worker. Workers can run on any installed harness, with an optional model selected for that task.

When harness is omitted, a worker inherits the master's harness. This makes a same-provider team the shortest path while still allowing mixed teams such as a Claude builder, Codex reviewer, and OpenCode validator.

Every worker appears on the canvas with its harness, role, activity, working-directory badge, and transcript visible. Hover the directory badge to see its full path.

The master is canvas-blind by default

A new master does not automatically read unrelated panes. It can list the canvas, but reading an existing pane requires an explicit association:

associate_pane({ pane_id })

After association, the master's read tools can inspect that pane. Associations persist per space, so a reopened master keeps the panes it already opted into while a new master starts blind.

Choose a harness and model

Both spawn_worker and spawn_role accept:

• harness - any installed worker harness, such as claude, codex, opencode, cursor, grok, devin, kimi, antigravity, or pi-mono.
• model - an optional harness-specific model ID. Omit it to use that harness's default or user configuration.

Call list_worker_harnesses before a cross-harness spawn when you need the current catalog and curated model suggestions. Model recommendations are guidance, not an exhaustive allowlist.

Worker sessions and resume

Workers are long-lived. Reuse an existing worker with send_to_worker when the follow-up belongs to the same role and context. This preserves its full conversation and avoids duplicating panes.

After a restart or master resume, use list_my_team to rediscover visible workers. Use list_recoverable_sessions to find recoverable sessions across installed harnesses, then pass the returned sessionId and harness to resume_worker.

Claude's interactive master sessions also appear under Sessions -> Resume Claude in the spawn menu, scoped to the active project folder.

Streaming and control

Worker panes stream text, reasoning, and tool activity as the turn runs. A master can read output, wait for one worker or the first of several workers, send a follow-up, and stop a worker. If a worker is interrupted, the transcript keeps the output already rendered.

You can rename any selected worker pane with f2f2, or ask the master to rename a pane it owns. Clear names make large review teams much easier to scan.

Usage indicator

The title bar shows usage for installed harnesses that expose usage data. AgentGrid only polls CLIs it detects, and the popover shows provider-specific windows, reset times, and stale or error states. Use Settings -> Agents to change harness order or accent colors.

The indicator is informational and never blocks a turn.

Next

• Coding harnesses - supported runtimes and provider-specific behavior.
• Orchestrating agents - spawn, wait, reuse, and review patterns.
• Roles - built-in worker roles and custom role configuration.
• System tray and daemon - what happens to workers when the window closes.