How to Use Muse Wonder in ChatGPT
Muse Wonder is built for multi-file code changes and tasks that require planning, testing, and revision. Maximum reasoning gives it more thinking time but does not turn it into Muse Wonder Full.
Prompts to try with Muse Wonder in ChatGPT
Copy an example into ChatGPT, or ask in your own words.
Use Muse Wonder with high reasoning to plan a six-week product launch. Ask for missing constraints, then give milestones, dependencies, and completion checks.
Use Muse Wonder with maximum reasoning to analyze this recurring software failure: a worker processes the same message twice after restarting. Rank hypotheses and propose tests that distinguish them.
Use Muse Wonder with high reasoning to turn the requirements in this conversation into a technical implementation plan. Flag contradictions and unresolved decisions.
Use Muse Wonder with high reasoning to revise our project plan after a two-week delay. Preserve the fixed deadline and explain which scope changes make it feasible.
Use Muse Wonder with maximum reasoning to compare the three migration approaches discussed here. Evaluate dependencies, failure recovery, and evidence needed before choosing.
How to set up Muse Wonder in ChatGPT
Pick the way that suits you. Each takes about two minutes.
Use the ChatGPT connector
- Add a custom MCP server
In ChatGPT open Plugins, then Add › Add custom MCP server. Name it Spiral, paste this URL and keep Authentication on OAuth. Tick I understand and want to continue, then click Create.
https://api.spiral.ai/mcp
- Click Connect
ChatGPT asks to connect Spiral. Connect takes you to Spiral in your browser.
- Sign in to Spiral and approve
Sign in, pick the organization ChatGPT should use and click Authorize. Spiral gives the connection its own API key; ChatGPT never sees it, and you can remove the connection under Settings › Connected apps.
Use the Spiral skill
- Send ChatGPT this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run Muse Wonder
One request; the result comes back by polling or a webhook.
curl https://api.spiral.ai/v1/runs \ -H "Authorization: Bearer $SPIRAL_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: $(uuidgen)" \ -d '{ "provider": "openrouter", "model": "meta/muse-wonder", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: ChatGPT finds Muse Wonder on Spiral, shows the price and runs it.
Muse Wonder providers and prices
Spiral runs Muse Wonder through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.106678 per run at the default inputs. $5.20 per 1M input tokens, $26.00 per 1M output tokens.
Settings you can ask for
Ask ChatGPT for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Top k | a whole number from 0 | ||
| Top p | 0 to 1 | ||
| Max tokens | 1 to 943718 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Reasoning effort | minimal, low, medium, high, xhigh, or max | medium | How much the model reasons. |
| Repetition penalty | 0 to 2 |
Every API field of Meta: Muse Wonder is on its tool page: open its card above.
What can you do with Muse Wonder in ChatGPT?
Muse Wonder can turn a complicated goal into a plan and revise it when new constraints arrive. Provide the goal, prior attempts, and completion criteria so the conversation stays focused on a useful result.
Refactor code across a repository
Use Muse Wonder to trace dependencies, change related modules, and verify that the revised code passes relevant tests.
Investigate a recurring bug
Have Muse Wonder compare hypotheses, reproduce the failure, and revise its diagnosis as test results arrive.
Plan a large code migration
Ask Muse Wonder to identify affected components, order the changes, and define validation and rollback criteria.
Turn a long brief into an execution plan
Give Muse Wonder the requirements and constraints, then request milestones, dependencies, and explicit completion checks.
Replan work after a blocker
Supply the current goal, completed work, failed attempts, and new constraints. Ask Muse Wonder for the next feasible approach.
Review a proposed technical change
Have Muse Wonder examine a proposal alongside relevant code and test results, then identify unresolved assumptions and verification work.
More on Muse Wonder and ChatGPT
Overview
Muse Wonder in other agents
Questions about Muse Wonder in ChatGPT
Can ChatGPT use Muse Wonder?
Yes. Add Spiral to ChatGPT as an MCP server, and ChatGPT can find and run Muse Wonder like any built-in tool.
How much does Muse Wonder cost in ChatGPT?
Up to $0.106678 per run at the default inputs. $5.20 per 1M input tokens, $26.00 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your ChatGPT plan is separate.
Do I need an OpenRouter account to use Muse Wonder in ChatGPT?
No. Spiral runs Muse Wonder on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up Muse Wonder in ChatGPT?
About two minutes: add a custom MCP server, then click Connect, then sign in to Spiral and approve.
Does ChatGPT wait while Muse Wonder runs?
Muse Wonder runs in the background. ChatGPT starts the run, checks on it and gives you the text, json when it's ready.
What is Muse Wonder best suited for?
Use it for code changes spanning multiple files and tasks that need repeated planning, testing, and revision. Meta emphasizes sustained work that adapts when an approach fails.
How much context can Muse Wonder handle?
The documented context window is 1,048,576 tokens. Input and the requested output budget share that space, and internal reasoning consumes part of the output budget.
Which reasoning setting should I use?
Meta recommends high reasoning for agentic coding. Use lower effort for simpler requests, and reserve maximum reasoning for difficult debugging or migration planning.
Does maximum reasoning give me Muse Wonder Full?
No. It increases the reasoning effort of Muse Wonder; it does not select Muse Wonder Full.
Where does Muse Wonder need extra checking?
Precise visual and spatial reasoning, particularly 3D tasks, is a documented weakness. Verify measurements and spatial conclusions against the original material.
Start using Muse Wonder in ChatGPT
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into ChatGPT:
set up https://spiral.ai/SKILL.md


