go

name	go
description	The single entry point to the Conductor system - state your goal and everything is handled automatically

/go -- Goal-Driven Entry Point

The single entry point to the entire Conductor system.

Just state your goal. The system handles everything else.

Usage

/go <your goal>

Examples

/go Add Stripe payment integration
/go Fix the login bug where users get logged out
/go Build a dashboard with analytics
/go Refactor the asset generation to use caching

What Happens

When you invoke /go, follow this process:

1. Goal Analysis

Parse the user's goal from $ARGUMENTS:

• Identify the type (feature, bugfix, refactor, etc.)
• Estimate complexity
• Extract key requirements

2. Track Detection

Check conductor/tracks.md for matching existing tracks:

• If match found: Resume that track from its current state
• If no match: Create a new track

3. For New Tracks

1. Create track directory: conductor/tracks/{goal-slug}_{date}/
2. Generate spec.md from the goal
3. Generate plan.md with DAG
4. Create metadata.json with v3 schema AND set superpower_enhanced: true (new tracks use superpowers by default)

Example metadata.json:

{
  "version": 3,
  "track_id": "goal-slug_20260213",
  "type": "feature",
  "status": "new",
  "superpower_enhanced": true,
  "loop_state": {
    "current_step": "NOT_STARTED",
    "step_status": "NOT_STARTED"
  }
}

4. Run the Evaluate-Loop

Invoke the conductor-orchestrator agent:

Use the conductor-orchestrator agent to run the evaluate-loop for this track.

The orchestrator will:

• Detect current step from metadata
• Check superpower_enhanced flag to determine which agents to use:
  • If true (new tracks): Dispatch superpowers (orchestrator-supaconductor:writing-plans, orchestrator-supaconductor:executing-plans, orchestrator-supaconductor:systematic-debugging)
  • If false/missing (legacy): Dispatch legacy loop agents (loop-planner, loop-executor, loop-fixer)
• Monitor progress and handle failures
• Complete the track or escalate if blocked

Decision Resolution

Behavior depends on conductor/config.json → "mode":

• "agentic" (default) — Fully autonomous. Never stops to ask the user. All decisions resolved by agents, leads, and board.
• "human-in-the-loop" — Pauses at decision points (ambiguity, blockers, fix limits, high-impact decisions) to ask the user.

See conductor/config.json for current mode. Edit to switch.

Resume Existing Work

/go                    # Continues the active track
/go continue           # Same as above