/parallel-dispatcher
The track ID to execute in parallel
> /plugin marketplace add Ibrahim-3d/orchestrator-supaconductor > /plugin install orchestrator-supaconductor@ibrahim-plugins
How it fires
How this command gets triggered: by you, by Claude, or both.
- Fires itselfClaude auto-loads it when your prompt matches the work.
- You can call itInvoke it directly when you want it.
- Slash command
/parallel-dispatcher
Context preview
What this command does when you run it.
The track ID to execute in parallel
Command definition
parallel-dispatcher.mdname: parallel-dispatcher
description: "Dispatch multiple task-workers in parallel based on DAG dependencies."
model: sonnet
arguments:
- name: track_id
description: "The track ID to execute in parallel"
required: true
user_invocable: true/orchestrator-supaconductor:parallel-dispatcher — Parallel Task Execution
Dispatch multiple task-workers simultaneously based on DAG dependencies.
Usage
/orchestrator-supaconductor:parallel-dispatcher <track-id>
Your Task
You ARE the parallel-dispatcher agent. Coordinate parallel execution.
1. Parse DAG from Plan
Read `conductor/tracks/{track_id}/plan.md` and extract the DAG section.
2. Initialize Message Bus
mkdir -p .message-bus/events
echo "{}" > .message-bus/locks.json
echo "{}" > .message-bus/worker-status.json3. Find Ready Tasks
Tasks are ready when all their `depends_on` tasks are complete.
4. Dispatch Workers in Parallel
Spawn multiple Claude sessions for conflict-free tasks:
# Spawn parallel workers (max 5 concurrent)
claude --print "/orchestrator-supaconductor:task-worker {track_id} 1.1" &
claude --print "/orchestrator-supaconductor:task-worker {track_id} 1.2" &
claude --print "/orchestrator-supaconductor:task-worker {track_id} 2.1" &
# Wait for batch to complete
wait5. Monitor Progress
# Check for completion events
ls .message-bus/events/TASK_COMPLETE_*.event
ls .message-bus/events/TASK_FAILED_*.event
6. Handle Failures
- Isolate failed tasks
- Continue with unblocked tasks
- Mark failed tasks for FIX step
7. Repeat Until Done
Find next batch of ready tasks and dispatch.
Worker Pool Limits
- Maximum 5 concurrent workers
- 30-minute timeout per worker
- Heartbeat required every 5 minutes
Output
PARALLEL EXECUTION COMPLETE
WORKERS SPAWNED: X
TASKS COMPLETED: Y
TASKS FAILED: Z
State Update
After all parallel groups complete:
- Set `current_step = "EVALUATE_EXECUTION"`
- Set `step_status = "NOT_STARTED"`
Message Bus
echo "PASS" > .message-bus/events/PARALLEL_COMPLETE_{track_id}.eventReference
Full agent instructions: `.claude/agents/orchestrator-supaconductor:parallel-dispatcher.md`
Read more
name: parallel-dispatcher
description: "Dispatch multiple task-workers in parallel based on DAG dependencies."
model: sonnet
arguments:
- name: track_id
description: "The track ID to execute in parallel"
required: true
user_invocable: true/orchestrator-supaconductor:parallel-dispatcher — Parallel Task Execution
Dispatch multiple task-workers simultaneously based on DAG dependencies.
Usage
/orchestrator-supaconductor:parallel-dispatcher <track-id>
Your Task
You ARE the parallel-dispatcher agent. Coordinate parallel execution.
1. Parse DAG from Plan
Read `conductor/tracks/{track_id}/plan.md` and extract the DAG section.
2. Initialize Message Bus
mkdir -p .message-bus/events
echo "{}" > .message-bus/locks.json
echo "{}" > .message-bus/worker-status.json3. Find Ready Tasks
Tasks are ready when all their `depends_on` tasks are complete.
4. Dispatch Workers in Parallel
Spawn multiple Claude sessions for conflict-free tasks:
# Spawn parallel workers (max 5 concurrent)
claude --print "/orchestrator-supaconductor:task-worker {track_id} 1.1" &
claude --print "/orchestrator-supaconductor:task-worker {track_id} 1.2" &
claude --print "/orchestrator-supaconductor:task-worker {track_id} 2.1" &
# Wait for batch to complete
wait5. Monitor Progress
# Check for completion events ls .message-bus/events/TASK_COMPLETE_*.event ls .message-bus/events/TASK_FAILED_*.event
6. Handle Failures
- Isolate failed tasks
- Continue with unblocked tasks
- Mark failed tasks for FIX step
7. Repeat Until Done
Find next batch of ready tasks and dispatch.
Worker Pool Limits
- Maximum 5 concurrent workers
- 30-minute timeout per worker
- Heartbeat required every 5 minutes
Output
PARALLEL EXECUTION COMPLETE WORKERS SPAWNED: X TASKS COMPLETED: Y TASKS FAILED: Z
State Update
After all parallel groups complete:
- Set `current_step = "EVALUATE_EXECUTION"`
- Set `step_status = "NOT_STARTED"`
Message Bus
echo "PASS" > .message-bus/events/PARALLEL_COMPLETE_{track_id}.eventReference
Full agent instructions: `.claude/agents/orchestrator-supaconductor:parallel-dispatcher.md`
Multi-agent orchestration system for Claude Code with parallel execution, automated quality gates, Board of Directors, and bundled Superpowers skills
Repo: Ibrahim-3d/orchestrator-supaconductor
Other commands on orchestrator-supaconductor.
- /board-meeting
The proposal, plan, or decision to deliberate on
Open command - /board-review
The proposal, plan, or decision to review
Open command - /brainstorm
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores requirements and design before implementation.
Open command - /brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores requirements and design before implementation.
Open command - /ceo
Your business question or topic to discuss
Open command - /close-track
Skip quality gate — use for abandoned or superseded tracks
Open command

