How to Use GPT-6 Luna in Amp
GPT-6 Luna suits repeatable document summaries, information extraction, and focused coding tasks. Choose it for everyday work; Luna Pro uses the same model with pro reasoning for harder problems.
Prompts to try with GPT-6 Luna in Amp
Copy an example into Amp, or ask in your own words.
Use GPT-6 Luna to summarize this repository's request validation flow. Cite the relevant files and functions, and keep the explanation under 400 words.
Use GPT-6 Luna to inspect the failing parser test, propose the smallest fix, and add a regression test. Report which tests actually ran.
Use GPT-6 Luna with high reasoning effort to review the database migration in this branch. Identify data-loss risks and suggest concrete validation queries.
Use GPT-6 Luna to extract the public function signatures from this package and draft concise reference documentation. Flag behavior the implementation does not establish.
Ask GPT-6 Luna through Spiral to review the diff on this branch and list the three riskiest changes. Save the result in ./assets.
How to set up GPT-6 Luna in Amp
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the server
No header and no client ID or secret: Amp registers itself with Spiral.
amp mcp add spiral https://api.spiral.ai/mcp
- Sign in
Start amp. Amp opens Spiral's sign-in page in your browser: sign in, pick the organization and click Authorize. No API key.
- Check it
amp mcp doctor
Use the Spiral skill
- Send Amp 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 GPT-6 Luna
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": "openai", "model": "openai/gpt-6-luna", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Amp finds GPT-6 Luna on Spiral, shows the price and runs it.
GPT-6 Luna providers and prices
Spiral runs GPT-6 Luna through OpenAI. Open a card for today's price, with every fee included.
Up to $0.002669 per run at the default inputs. $0.13 per 1M input tokens, $0.013 per 1M cached input tokens, $0.1625 per 1M cache write tokens, $0.65 per 1M output tokens.
Settings you can ask for
Ask Amp for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Top p | a number up to 1 | Taken only with reasoning effort "none". | |
| Logprobs | on or off | Taken only with reasoning effort "none". | |
| Max tokens | 16 to 128000 | The most output tokens to generate. | |
| Temperature | 0 to 2 | Taken only with reasoning effort "none". | |
| Top logprobs | 0 to 20 | Taken only with reasoning effort "none". | |
| Presence penalty | -2 to 2 | Taken only with reasoning effort "none". | |
| Reasoning effort | none, low, medium, high, xhigh, or max | medium | How much the model reasons. |
| Frequency penalty | -2 to 2 | Taken only with reasoning effort "none". | |
| Parallel tool calls | on or off |
Every API field of GPT-6 Luna is on its tool page: open its card above.
What can you do with GPT-6 Luna in Amp?
Turn project files into focused explanations, patches, and testable debugging hypotheses. Use Luna for bounded tasks and Luna Pro when the problem needs closer examination.
Extract facts from repeated documents
Use Luna to pull dates, commitments, and named parties from a set of documents. Request a supporting passage for each extracted item and mark missing information explicitly.
Turn long material into a focused brief
Ask Luna to summarize decisions, unresolved questions, and next actions. Specify the audience and a word limit so the result answers a concrete need.
Investigate a difficult software defect
Use Luna Pro to examine competing explanations for a failing test. Request a reproducible case, a proposed fix, and evidence that distinguishes the explanations.
Review a complicated plan
Ask Luna Pro to inspect a migration or rollout plan for conflicting assumptions, missing dependencies, and failure cases. Require each finding to reference the supplied plan.
Read information from an image
Give Luna an image containing useful text or visual information. Ask for a bounded extraction and have it flag anything it cannot read confidently.
Prepare recurring operational summaries
Have a scheduled agent send Luna the latest updates and request a short summary of changes, unresolved issues, and questions requiring human attention.
More on GPT-6 Luna and Amp
Overview
GPT-6 Luna in other agents
Questions about GPT-6 Luna in Amp
Can Amp use GPT-6 Luna?
Yes. Add Spiral to Amp as an MCP server, and Amp can find and run GPT-6 Luna like any built-in tool.
How much does GPT-6 Luna cost in Amp?
Up to $0.002669 per run at the default inputs. $0.13 per 1M input tokens, $0.013 per 1M cached input tokens, $0.1625 per 1M cache write tokens, $0.65 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Amp plan is separate.
Do I need an OpenAI account to use GPT-6 Luna in Amp?
No. Spiral runs GPT-6 Luna on its own OpenAI account. You sign in to Spiral once and never paste a key.
How long does it take to set up GPT-6 Luna in Amp?
About two minutes: add the server, then sign in, then check it.
Does Amp wait while GPT-6 Luna runs?
GPT-6 Luna runs in the background. Amp starts the run, checks on it and gives you the text, json when it's ready.
What is GPT-6 Luna best used for?
Start with repeatable work: document summaries, extracting information, and answering focused questions. It also produces code and can tackle software engineering tasks.
How is GPT-6 Luna Pro different from Luna?
Both use the same underlying model. Luna Pro enables pro reasoning for complex tasks, making it a candidate for difficult debugging or reviewing a complicated plan.
How much text can GPT-6 Luna handle and generate?
Luna has a 1,050,000-token context window and a maximum output of 128,000 tokens. Reasoning also consumes the generation budget, so leave room for it when requesting a long answer.
Can GPT-6 Luna understand images?
Yes. It accepts images and text and produces text, including code, so you can ask it to extract information from an image or explain what it shows.
How should I prompt GPT-6 Luna?
Give it a clear task, the relevant source material, and a specific definition of a successful answer. OpenAI recommends brief, direct instructions for reasoning models; asking for step-by-step internal reasoning is unnecessary.
Start using GPT-6 Luna in Amp
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Amp:
set up https://spiral.ai/SKILL.md


