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/faq.md

View on GitHub
# FAQ

Part of the `multi-model-orchestrator` skill. Common questions. Load this when
something about the workflow is unclear.

**Q: Can I use this with Claude, Opus, Haiku, Grok, Mistral, etc.?**

A: Yes. The skill is agent-agnostic. Use any LLM model. The `executor` field is just a label.

**Q: Should I commit handoffs to git?**

A: Commit handoffs only when the repository is private or the file has been
reviewed for secrets, customer data, internal plans, and copied prompts/logs.
For sensitive work, keep handoffs outside git or commit a redacted summary.

**Q: Can I parallelize execution?**

A: Yes. If subtasks have no dependencies, execute them simultaneously. Record results when they finish.

**Q: What if a subtask blocks?**

A: Record it in `execution.rounds` with `status: blocked`. Decide whether to:
1. Fix the blocker and retry
2. Reassign to a different agent
3. Escalate to human review

**Q: How long should a subtask be?**

A: Aim for 15-60 minute chunks. If a subtask would take 4+ hours, break it down further.

**Q: Can I have multiple handoffs active?**

A: Yes. Use separate files: `task-1.yaml`, `task-2.yaml`. Each one is independent.

**Q: How is this different from running agents sequentially?**

A: This skill structures the work so:
- Multiple agents can work in parallel
- Results are centralized in one document
- Changes and decisions are all traceable
- You can review and optimize the overall plan, not just individual agent outputs
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