Claude Code Plugins

Community-maintained marketplace

Feedback
20
0

Multi-agent parallel execution with autonomous workers. Use when plan.md has Track Assignments section or user triggers /conductor-orchestrate, "run parallel", "spawn workers". MUST load maestro-core skill first for routing.

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md

name orchestrator
description Multi-agent parallel execution with autonomous workers. Use when plan.md has Track Assignments section or user triggers /conductor-orchestrate, "run parallel", "spawn workers". MUST load maestro-core skill first for routing.

Orchestrator - Multi-Agent Parallel Execution

Spawn autonomous workers to execute tracks in parallel using Agent Mail coordination.

Core Principles

  • Load core first - Load maestro-core for routing table and fallback policies
  • Pre-register workers before spawning (Agent Mail validates recipients)
  • Workers own their beads - can bd claim/close directly (unlike sequential mode)
  • File reservations prevent conflicts - reserve before edit, release on complete
  • Summary before exit - all workers MUST send completion message
  • TDD by default - workers follow RED → GREEN → REFACTOR cycle (use --no-tdd to disable)

When to Use

Trigger Condition
Auto-routed /conductor-implement when plan has Track Assignments
File-scope /conductor-implement when ≥2 non-overlapping file groups detected
Direct /conductor-orchestrate or co
Phrase "run parallel", "spawn workers", "dispatch agents"

Confirmation Prompt

Before spawning parallel workers, orchestrator displays a confirmation:

📊 Parallel execution detected:
- Track A: 2 tasks (src/api/)
- Track B: 2 tasks (lib/)
- Track C: 1 task (schemas/)

Run parallel? [Y/n]:

See implement.md Phase 2b for full confirmation flow.

Quick Reference

Action Tool
Parse plan.md Read("conductor/tracks/<id>/plan.md")
Initialize ensure_project, register_agent
Spawn workers Task() for each track
Monitor fetch_inbox, search_messages
Resolve blockers reply_message
Complete Send summary, bd close epic
Track threads summarize_thread(thread_id=TRACK_THREAD)
Auto-routing Auto-detect parallel via metadata.json.beads

8-Phase Workflow

  1. Preflight - Session identity, detect active sessions
  2. Read Plan - Parse Track Assignments from plan.md
  3. Validate - Health check Agent Mail (HALT if unavailable)
  4. Initialize - ensure_project, register orchestrator + all workers
  5. Spawn Workers - Task() for each track (parallel)
  6. Monitor + Verify - fetch_inbox, verify worker summaries
    • Workers use track threads (TRACK_THREAD) for bead-to-bead context
  7. Resolve - reply_message for blockers
  8. Complete - Send summary, close epic, rb review

See references/workflow.md for full protocol.

Worker 4-Step Protocol

All workers MUST follow this exact sequence:

┌─────────────────────────────────────────────────────────────┐
│  STEP 1: INITIALIZE  - macro_start_session() FIRST         │
│  STEP 2: EXECUTE     - claim beads, do work, close beads   │
│  STEP 3: REPORT      - send_message() to orchestrator      │
│  STEP 4: CLEANUP     - release_file_reservations()         │
└─────────────────────────────────────────────────────────────┘
Step Tool Required
1 macro_start_session() ✅ FIRST
2 bd update, bd close
3 send_message() ✅ LAST
4 release_file_reservations()

Critical rules:

  • ❌ Never start work before macro_start_session()
  • ❌ Never return without send_message() to orchestrator
  • ❌ Never touch files outside assigned scope

See references/worker-prompt.md for full template.

Agent Routing

Intent Keywords Agent Type File Reservation
research, find, locate Research None (read-only)
review, check, audit Review None (read-only)
plan, design, architect Planning conductor/tracks/**
implement, build, create Execution Task-specific scope
fix, debug, investigate Debug None (read-only)

See references/intent-routing.md for mappings.

Anti-Patterns

❌ Don't ✅ Do
Spawn workers without pre-registration Register all workers BEFORE spawning
Skip completion summary Always send_message before exit
Ignore file reservation conflicts Wait or resolve before proceeding
Use orchestration for simple tasks Use sequential /conductor-implement

Lazy References

Load references only when needed:

Phase Trigger Condition Reference
Always On skill load SKILL.md (this file)
Phase 3 (Initialize) Setting up Agent Mail, project registration agent-mail.md
Phase 4 (Spawn) Before dispatching worker agents worker-prompt.md
Phase 6 (Handle Issues) Cross-track dependencies, blocker resolution agent-coordination.md

All References

Topic File
Full workflow workflow.md
Architecture architecture.md
Coordination modes coordination-modes.md
Agent Mail protocol agent-mail.md
Worker prompt template worker-prompt.md
Preflight/session brain preflight.md
Intent routing intent-routing.md
Summary format summary-protocol.md
Auto-routing auto-routing.md
Track threads track-threads.md

Related

  • maestro-core - Routing and fallback policies
  • conductor - Track management, /conductor-implement
  • beads - Issue tracking, bd commands