How to Use o3 in Cursor
Use o3 to debug difficult code, solve math problems, and interpret charts. Compared with o3-pro, it suits complex work that does not need the extra deliberation aimed at more reliable answers.
Prompts to try with o3 in Cursor
Copy an example into Cursor, or ask in your own words.
Use o3 to investigate the failing tests in this project. Inspect the relevant code, identify the likely cause, and propose the smallest patch with a regression test.
Use o3 to compare the implementation with the requirements in this repository. Report concrete mismatches, the affected files, and a test that would expose each mismatch.
Use o3 to plan the refactor described in the issue. Inspect project dependencies and propose an ordered sequence of changes, with a verification checkpoint after each change.
Ask o3 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 o3 on this function, then tell me what the call cost.
How to set up o3 in Cursor
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the hosted server
Use Add to Cursor on the console's Agents & MCP page, or add this to .cursor/mcp.json (one project) or ~/.cursor/mcp.json (every project):
{ "mcpServers": { "spiral": { "url": "https://api.spiral.ai/mcp" } } } - Sign in
Open Cursor Settings › MCP, click Connect next to spiral, sign in to Spiral and click Authorize. No API key. Spiral then shows as connected, with its tools listed.
Use the Spiral skill
- Send Cursor 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 o3
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/o3", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Cursor finds o3 on Spiral, shows the price and runs it.
o3 providers and prices
Spiral runs o3 through OpenAI. Open a card for today's price, with every fee included.
Up to $0.04269 per run at the default inputs. $2.60 per 1M input tokens, $0.65 per 1M cached input tokens, $10.40 per 1M output tokens.
Settings you can ask for
Ask Cursor for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Max tokens | 16 to 100000 | The most output tokens to generate. | |
| Parallel tool calls | on or off |
Every API field of o3 is on its tool page: open its card above.
What can you do with o3 in Cursor?
OpenAI o3 can help turn a difficult bug or design question into a concrete patch or implementation plan. Give it relevant project code, constraints, and test results, then request a result you can verify.
Investigate a difficult bug
Give OpenAI o3 the failing behavior, relevant code, and test results. Ask for a likely cause, a minimal fix, and a regression test that demonstrates the problem.
Review a database migration
Have OpenAI o3 compare a proposed migration with its requirements. Request failure scenarios, data consistency checks, and a rollback plan.
Check a mathematical solution
Supply the problem and proposed answer to OpenAI o3. Ask it to identify invalid assumptions, check boundary cases, and return a corrected solution when needed.
Compare scientific explanations
Give OpenAI o3 observations and competing hypotheses. Request the evidence for each explanation and an experiment that could distinguish them.
Interpret a chart
Provide OpenAI o3 with a chart image and its surrounding context. Ask for the main trend, supporting observations, and claims the chart does not establish.
Compare conflicting requirements
Ask OpenAI o3 to reconcile supplied specifications or policy excerpts. Request the conflicting passages, their practical consequences, and the decisions still needed.
Plan a complicated project
Give OpenAI o3 the goal, constraints, and dependencies. Ask for an ordered plan with checkpoints and explicit conditions for moving to the next task.
Find evidence across long source material
Provide OpenAI o3 with labeled source excerpts and a specific question. Ask it to extract relevant passages and distinguish documented conclusions from unresolved questions.
More on o3 and Cursor
Overview
o3 in other agents
Questions about o3 in Cursor
Can Cursor use o3?
Yes. Add Spiral to Cursor as an MCP server, and Cursor can find and run o3 like any built-in tool.
How much does o3 cost in Cursor?
Up to $0.04269 per run at the default inputs. $2.60 per 1M input tokens, $0.65 per 1M cached input tokens, $10.40 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Cursor plan is separate.
Do I need an OpenAI account to use o3 in Cursor?
No. Spiral runs o3 on its own OpenAI account. You sign in to Spiral once and never paste a key.
How long does it take to set up o3 in Cursor?
About two minutes: add the hosted server, then sign in.
Does Cursor wait while o3 runs?
o3 runs in the background. Cursor starts the run, checks on it and gives you the text, json when it's ready.
What is OpenAI o3 good at?
OpenAI o3 is suited to difficult coding, math, science, and visual analysis. Useful tasks include investigating a bug, checking a calculation, or interpreting a chart alongside its supporting explanation.
How does o3-pro differ from o3?
o3-pro uses o3 as its underlying model and thinks longer to produce more reliable responses. Choose it for a difficult question where you want additional deliberation.
How much material can OpenAI o3 handle?
o3 and o3-pro each have a 200,000-token context window and a maximum output allowance of 100,000 tokens. Internal reasoning also uses generated tokens, so that allowance does not mean 100,000 tokens of readable text.
How should I prompt OpenAI o3?
Use brief, direct instructions. State the question, provide the relevant evidence, and specify the final result you need, such as a patch, comparison table, or short recommendation.
Does OpenAI o3 know about recent events?
o3 and o3-pro have a June 1, 2024 knowledge cutoff. For later events, provide dated source material and ask the model to base its answer on that evidence.
Start using o3 in Cursor
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Cursor:
set up https://spiral.ai/SKILL.md


