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/handoff-structure.md

View on GitHub
# Handoff Structure Reference

Part of the `multi-model-orchestrator` skill. Field-by-field reference for the
handoff YAML. Load this when you need to recall an exact field name, type, or
purpose.

### `metadata`

| Field | Type | Required | Purpose |
|-------|------|----------|---------|
| `name` | string | yes | Human-readable task name |
| `created` | date | yes | When the handoff was created |
| `goal` | string | yes | Summary of what we are doing |
| `status` | enum | yes | planning \| executing \| review \| complete |

### `goal`

| Field | Type | Purpose |
|-------|------|---------|
| `summary` | string | Detailed description of the goal |
| `acceptance_criteria` | array | Testable completion conditions |
| `context` | string | Background, constraints, prior decisions |

### `subtasks`

| Field | Type | Purpose |
|-------|------|---------|
| `id` | string | Unique identifier (e.g. task-1, task-auth-setup) |
| `title` | string | Short, clear description |
| `type` | enum | code \| analysis \| design \| research \| validation |
| `executor` | string | Which agent: Opus, Haiku, Codex, Claude, Claude Code |
| `input` | string | Everything the executor needs to know |
| `acceptance` | string | How to verify completion |
| `depends_on` | array | List of task IDs this depends on |
| `status` | enum | pending \| in-progress \| done \| blocked |

### `execution.rounds`

| Field | Type | Purpose |
|-------|------|---------|
| `round` | number | Iteration number |
| `task_id` | string | Which subtask was executed |
| `executor` | string | Which agent did it |
| `status` | enum | done \| blocked \| partial |
| `result` | string | What was delivered |
| `issues` | string | Blockers or concerns |
| `next_step` | string | Recommendation for next action |
| `timestamp` | ISO 8601 | When it was executed |

### `feedback`

| Field | Type | Purpose |
|-------|------|---------|
| `synthesis` | string | Overall assessment: what worked, what didn't, what changed |
| `improvements` | array | List of ideas for next iteration |
| `decision_log` | string | Why we made certain choices and tradeoffs |
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: 472kUpdated 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