Spiral gives your agent every tool with one account. Browse the tools →

Factory Droid · LLMReleased 2026-10-12

How to Use Muse Wonder in Factory Droid

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 Factory Droid

Copy an example into Factory Droid, or ask in your own words.

Use Muse Wonder with high reasoning to trace the authentication flow in this repository. Identify inconsistent permission checks, fix them, and run the relevant tests.

Muse WonderFactory Droid
Try asking
  • Use Muse Wonder with maximum reasoning to investigate duplicate queue processing after a restart. Reproduce the failure, compare likely causes, and verify the smallest fix.

  • Use Muse Wonder with high reasoning to refactor duplicated validation across this project. Preserve behavior and verify each affected entry point.

  • Use Muse Wonder with maximum reasoning to plan this repository's database migration. Identify dependencies, rollback conditions, and tests before making changes.

  • Use Muse Wonder with high reasoning to review the current diff against the project's requirements. Investigate failures and report which acceptance criteria remain unverified.

How to set up Muse Wonder in Factory Droid

Pick the way that suits you. Each takes about two minutes.

Use Spiral's MCP server

  1. Add the server

    Droid saves it to your user config, ~/.factory/mcp.json, so it works in every project.

    droid mcp add spiral https://api.spiral.ai/mcp --type http
  2. Sign in

    Start droid, type /mcp, select spiral and authenticate it. Sign in to Spiral in the browser and click Authorize. No API key.

  3. Check it

    Spiral shows as connected. needs authentication means the sign-in did not finish: run /mcp again.

    droid mcp list

Use the Spiral skill

  1. Send Factory Droid this message

    Your agent reads Spiral's skill and sets itself up.

    set up https://spiral.ai/SKILL.md
  2. Sign in or paste a key

    Approve Spiral in the browser, or give the agent a key from the console.

Use the Spiral CLI

  1. 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

  1. Create an API key

    In the console, under API keys.

  2. 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: Factory Droid 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.

OpenRouterLLM
Meta: Muse Wonder
meta/muse-wonder

Settings you can ask for

Ask Factory Droid for any of these in your prompt.

SettingAllowed valuesDefaultDescription
Top ka whole number from 0
Top p0 to 1
Max tokens1 to 943718

The most output tokens to generate.

Temperature0 to 2
Reasoning effortminimal, low, medium, high, xhigh, or maxmedium

How much the model reasons.

Repetition penalty0 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 Factory Droid?

Muse Wonder can carry a repository change through investigation, edits, and verification across multiple files. Give it acceptance criteria and ask it to use test results to guide revisions.

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.

Questions about Muse Wonder in Factory Droid

Can Factory Droid use Muse Wonder?

Yes. Add Spiral to Factory Droid as an MCP server, and Factory Droid can find and run Muse Wonder like any built-in tool.

How much does Muse Wonder cost in Factory Droid?

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 Factory Droid plan is separate.

Do I need an OpenRouter account to use Muse Wonder in Factory Droid?

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 Factory Droid?

About two minutes: add the server, then sign in, then check it.

Does Factory Droid wait while Muse Wonder runs?

Muse Wonder runs in the background. Factory Droid 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.

Get started

Start using Muse Wonder in Factory Droid

One Spiral account for every tool your agent needs. No subscription; you pay per run.

Paste this into Factory Droid:

Set up Factory Droid

set up https://spiral.ai/SKILL.md
Are you an AI agent? See setup options