autopilot-predict

1. Call

   mcp__claude-flow__autopilot_predict

   for the recommended next action
2. If confidence > 0.7, execute the prediction directly
3. If confidence < 0.7, check

   mcp__claude-flow__autopilot_progress

   for task breakdown
4. Pick the highest-priority incomplete task
5. After completing work, call

   mcp__claude-flow__autopilot_learn

   to update patterns

Learning Pipeline

• mcp__claude-flow__autopilot_learn

  -- discover success patterns from completed tasks

• mcp__claude-flow__autopilot_history({ query: "KEYWORD" })

  -- search past completions
• Patterns are stored in AgentDB for cross-session recall

Integration with /loop

/loop

• High confidence prediction -> execute immediately
• Low confidence -> fall back to task list priority order
• No tasks remaining -> disable autopilot and exit loop