How to Use GPT-6 Astra in Codex
GPT-6 Astra handles complex coding, research, and document work. Choose it for regular agent tasks; Pro uses the same underlying model with additional reasoning for harder problems.
Prompts to try with GPT-6 Astra in Codex
Copy an example into Codex, or ask in your own words.
Use GPT-6 Astra to investigate the failing checkout test in this repository. Trace the cause, implement the smallest complete fix, and run the affected tests. Finish with the cause, change, and verification result.
Use GPT-6 Astra to implement keyboard navigation for the existing settings dialog. Follow the project's conventions, verify focus movement and dismissal, and continue until the requested behavior works.
Use GPT-6 Astra to inspect the repository's instructions for redundant testing requirements. Propose targeted edits that preserve necessary checks and tie each requirement to the changes it actually covers.
Ask GPT-6 Astra through Spiral to review the diff on this branch and list the three riskiest changes. Save the result in ./assets.
Get a second opinion from GPT-6 Astra on this function, then tell me what the call cost.
How to set up GPT-6 Astra in Codex
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the hosted server
codex mcp add spiral --url https://api.spiral.ai/mcp
- Sign in
Codex opens the browser: sign in to Spiral and click Authorize. No API key. codex mcp list then shows Spiral.
codex mcp login spiral
Use the Spiral skill
- Send Codex 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 Astra
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-astra", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Codex finds GPT-6 Astra on Spiral, shows the price and runs it.
GPT-6 Astra providers and prices
Spiral runs GPT-6 Astra through OpenAI. Open a card for today's price, with every fee included.
Up to $0.26681 per run at the default inputs. $13.00 per 1M input tokens, $1.30 per 1M cached input tokens, $16.25 per 1M cache write tokens, $65.00 per 1M output tokens.
Settings you can ask for
Ask Codex for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Max tokens | 16 to 128000 | The most output tokens to generate. | |
| Reasoning effort | low, medium, high, xhigh, or max | How much the model reasons. | |
| Parallel tool calls | on or off |
Every API field of GPT-6 Astra is on its tool page: open its card above.
What can you do with GPT-6 Astra in Codex?
Turn a bug report or technical requirement into an implemented, checked change. Define the affected behavior and completion criteria so Astra can carry the work through.
Investigate and fix a software defect
Ask Astra to trace a reproducible failure, implement a focused fix, and verify the affected behavior. Define the stopping point as a working result with a short explanation of the evidence.
Review a difficult technical decision
Give Pro competing designs, requirements, and known constraints. Request a recommendation that names failure modes, unresolved assumptions, and evidence that could change the decision.
Turn research into a decision memo
Provide source material and a concrete question. Request a concise memo that separates supported conclusions, conflicting evidence, and open questions.
Check a website's behavior
With browser tools available, ask Astra to exercise a specific user journey, record reproducible failures, and verify corrections.
Review a scientific analysis
Provide observations, methods, and the question under investigation. Ask Astra to inspect the calculations, identify competing explanations, and suggest a test that distinguishes them.
Create a report from source material
Specify the audience, required sections, and length. Ask Astra to produce the report and identify any conclusions the supplied evidence cannot support.
More on GPT-6 Astra and Codex
Overview
GPT-6 Astra in other agents
Questions about GPT-6 Astra in Codex
Can Codex use GPT-6 Astra?
Yes. Add Spiral to Codex as an MCP server, and Codex can find and run GPT-6 Astra like any built-in tool.
How much does GPT-6 Astra cost in Codex?
Up to $0.26681 per run at the default inputs. $13.00 per 1M input tokens, $1.30 per 1M cached input tokens, $16.25 per 1M cache write tokens, $65.00 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Codex plan is separate.
Do I need an OpenAI account to use GPT-6 Astra in Codex?
No. Spiral runs GPT-6 Astra 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 Astra in Codex?
About two minutes: add the hosted server, then sign in.
Does Codex wait while GPT-6 Astra runs?
GPT-6 Astra runs in the background. Codex starts the run, checks on it and gives you the text, json when it's ready.
What is GPT-6 Astra good at?
It is built for complex reasoning, coding, research, computer use, and document creation. Useful assignments include investigating a bug, comparing research evidence, or turning source material into a finished report.
How does GPT-6 Astra Pro differ from GPT-6 Astra?
Pro uses the same underlying model in Pro reasoning mode, with additional model work aimed at higher-quality answers to complex tasks. Use regular Astra for routine work and consider Pro for a difficult analysis or review.
How much material can GPT-6 Astra handle?
The documented context window is 1,050,000 tokens, with up to 922,000 input tokens and 128,000 output tokens. Reasoning also consumes the generated-token allowance, so 128,000 tokens does not mean an answer of that length.
Can GPT-6 Astra understand images or generate media?
Its native inputs are text and images, and its native output is text. Use image input for analysis; generating image, audio, or video files requires separate tools or models.
Why does GPT-6 Astra sometimes stop before finishing?
OpenAI documents a tendency to ask for clarification where users expect reasonable assumptions. State what counts as finished, which decisions it can make independently, and which missing details actually require a question.
Start using GPT-6 Astra in Codex
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Codex:
set up https://spiral.ai/SKILL.md


