Muse Code Verified
Key insights
Concrete technical or product signals.
- Muse Wonder Full is the reason to pick a Power or Maximum Usage subscription; it has no Model API ID.
- Muse Wonder through pay-as-you-go costs $4 input, $0.20 cached input, and $20 output per million tokens.
- Use Muse Code when you want Meta's ready-made harness; call Meta Model API directly when you are building your own agent.
Use cases
Where this shines in production.
- Interactive repository coding with approvals and a sandbox
- Headless CI runs via muse exec
- Long-running engineering tasks on Muse Wonder Full
Limitations & trade-offs
What to watch for.
- Official install is macOS and Linux; do not assume Windows support from the docs.
- Muse Wonder Full requires the Power Usage or Maximum Usage plan.
- A local coding agent still needs secret hygiene, tests, branch protection, and human review.
Models referenced
Declared model dependencies or integrations.
Muse Wonder Full, Muse Wonder 1.0, Muse Spark 1.3
Related prompts
Hand-picked or latest prompt templates.
Prompt
Prompt-injection resistant assistant
System prompt that treats retrieved docs, tickets, and web pages as untrusted data. Use in multi-tenant RAG and browsing agents. This is a defensive template, not an attack or bypass playbook.
Prompt
Computer-use agent with approval gates
System prompt for browser/desktop agents (Grok Bot, computer-use tools). Allows research and drafts, but pauses before send, purchase, delete, or production changes. Not a jailbreak or bypass guide.
Prompt
Transcript to summary and action items
Turn a speech transcript into decisions, owners, and follow-ups. Use after Gemini 3.5 Transcribe, Whisper, or MAI-Transcribe. Do not use on raw audio; transcribe first.
Prompt
Product image generation prompt
A reusable image prompt for catalog and hero shots: subject, lighting, lens, negative constraints, and text-in-image rules. Works with Firefly, Midjourney, MAI-Image, Qwen-Image, and SDXL. Not for photorealistic people or brand-logo cloning.
Prompt
LLM output evaluation rubric
Judge a model answer against a task, retrieved evidence, and pass/fail thresholds. Use for golden-set grading and regression evals. Not a substitute for human review on high-risk decisions.
Prompt
Strict JSON extraction prompt
Extract structured fields into JSON that matches a schema. Pair with provider structured-output / JSON-schema modes when available. Do not use when the output is meant to be prose.
Looking for a tighter match? Search the prompt library.
Muse Code FAQ
What is Muse Code?
Muse Code is Meta's coding agent for the terminal and CI. The official installer places a muse binary on macOS and Linux. Interactive sessions plan, edit, and run commands with workspace trust, approvals, and an OS sandbox, and headless muse exec runs one prompt for scripts and CI. Since October 12, 2026 it can run Muse Wonder (muse-wonder-1.0), Meta's frontier agent and coding model, and Muse Wonder Full, which is only available inside Muse Code on the Power Usage plan ($50/month, 20x Everyday usage) and the new Maximum Usage plan ($200/month, 80x Everyday usage). Muse Spark 1.3 remains available. Billing is Meta Model API pay-as-you-go or a Muse Code monthly subscription.
When should teams use Muse Code?
Interactive repository coding with approvals and a sandbox
What should teams watch out for with Muse Code?
Official install is macOS and Linux; do not assume Windows support from the docs.
Related
Comparisons, platforms, and models teams often view next.