Skip to content

majiayu000/spellbook / skills/multi-model-orchestrator

multi-model-orchestrator

Verified

Use when coordinating complex tasks across multiple AI agents with a centralized handoff document for planning, execution tracking, and feedback fusion.

references/workflow-patterns.md

View on GitHub
# Workflow Patterns

Part of the `multi-model-orchestrator` skill. Common subtask dependency
layouts. Load this when you are designing the `depends_on` graph for a handoff.

## Pattern 1: Parallel Execution

All subtasks with no dependencies run simultaneously:

```yaml
subtasks:
  - id: task-1
    title: "Analyze requirements"
    executor: "Claude"
    depends_on: []

  - id: task-2
    title: "Design architecture"
    executor: "Opus"
    depends_on: []  # No dependency on task-1

  - id: task-3
    title: "Set up project"
    executor: "Claude Code"
    depends_on: []  # Independent
```

→ Run all three agents at the same time, then sync results.

## Pattern 2: Sequential with Dependency Chain

Each task waits for the previous one:

```yaml
subtasks:
  - id: task-1
    title: "Understand the problem"
    executor: "Fable"
    depends_on: []

  - id: task-2
    title: "Write the code"
    executor: "Codex"
    depends_on: ["task-1"]  # Waits for task-1

  - id: task-3
    title: "Test and validate"
    executor: "Claude Code"
    depends_on: ["task-2"]  # Waits for task-2
```

→ Run sequentially, using output of each as input to the next.

## Pattern 3: Parallel with Sync Point

Some tasks run in parallel; others converge:

```yaml
subtasks:
  - id: task-1a
    executor: "Opus"
    depends_on: []

  - id: task-1b
    executor: "Haiku"
    depends_on: []

  - id: task-2
    title: "Synthesize results"
    executor: "Claude"
    depends_on: ["task-1a", "task-1b"]  # Waits for both
```

→ 1a and 1b run in parallel; 2 waits for both to finish.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall