How to Use Claude Opus 5 in Cursor
Claude Opus 5 suits difficult debugging, large refactors, and tasks that span many steps. Compared with Claude Opus 4.8, its coding improvements are most useful for complex changes across multiple files.
Prompts to try with Claude Opus 5 in Cursor
Copy an example into Cursor, or ask in your own words.
Use Claude Opus 5 to trace the failing parser test to its cause. Fix the affected code while preserving the public interface.
Use Claude Opus 5 to consolidate duplicated validation across this project. Preserve existing error messages and keep the changes limited to validation.
Use Claude Opus 5 to implement pagination across the endpoint and client. Follow the project's conventions and return a concise change summary.
Use Claude Opus 5 with xhigh reasoning effort to investigate the intermittent cache failure. Address the cause and explain the resulting behavior in 150 words.
Use Claude Opus 5 to build an exchange-message parser from the specification in this repository. Cover the documented message types and malformed input behavior.
How to set up Claude Opus 5 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 Claude Opus 5
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": "anthropic", "model": "anthropic/claude-opus-5", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Cursor finds Claude Opus 5 on Spiral, shows the price and runs it.
Claude Opus 5 providers and prices
Spiral runs Claude Opus 5 through Anthropic. Open a card for today's price, with every fee included.
Up to $0.133575 per run at the default inputs. $6.50 per 1M input tokens, $0.65 per 1M cached input tokens, $8.125 per 1M cache write tokens, $13.00 per 1M 1-hour cache write tokens, $32.50 per 1M output tokens; for a prompt above 200,000 tokens, $13.00 per 1M input tokens, $1.30 per 1M cached input tokens, $16.25 per 1M cache write tokens, $26.00 per 1M 1-hour cache write tokens, $48.75 per 1M output tokens.
Settings you can ask for
Ask Cursor for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Max tokens | 1 to 128000 | The most output tokens to generate. On a reasoning model the reasoning tokens count toward it: a cap too low for the effort can end with no answer (finish_reason length). | |
| Reasoning effort | low, medium, high, xhigh, or max | high | How much the model reasons. |
| Parallel tool calls | on or off |
Every API field of Claude Opus 5 is on its tool page: open its card above.
What can you do with Claude Opus 5 in Cursor?
Claude Opus 5 can carry a difficult bug fix or feature across several project files. Give it the intended behavior and scope so the changes stay focused.
Fix bugs that survive an initial patch
Use Claude Opus 5 to trace a failure to its cause and address related edge cases.
Refactor code across multiple files
Use Claude Opus 5 for coordinated changes across modules while preserving the behavior you specify.
Implement a complete feature
Give Claude Opus 5 the behavior, constraints, and acceptance criteria for a feature spanning several parts of an application.
Compare lengthy reports
Use Claude Opus 5 to compare supplied reports, surface disagreements, and link conclusions to supporting passages.
Develop spreadsheet formulas
Ask Claude Opus 5 to design formulas across related sheets and explain their assumptions.
Review account health
Use Claude Opus 5 to turn account-health data into a prioritized risk summary and suggested next actions.
More on Claude Opus 5 and Cursor
Claude Opus 5 in other agents
Questions about Claude Opus 5 in Cursor
Can Cursor use Claude Opus 5?
Yes. Add Spiral to Cursor as an MCP server, and Cursor can find and run Claude Opus 5 like any built-in tool.
How much does Claude Opus 5 cost in Cursor?
Up to $0.133575 per run at the default inputs. $6.50 per 1M input tokens, $0.65 per 1M cached input tokens, $8.125 per 1M cache write tokens, $13.00 per 1M 1-hour cache write tokens, $32.50 per 1M output tokens; for a prompt above 200,000 tokens, $13.00 per 1M input tokens, $1.30 per 1M cached input tokens, $16.25 per 1M cache write tokens, $26.00 per 1M 1-hour cache write tokens, $48.75 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 Anthropic account to use Claude Opus 5 in Cursor?
No. Spiral runs Claude Opus 5 on its own Anthropic account. You sign in to Spiral once and never paste a key.
How long does it take to set up Claude Opus 5 in Cursor?
About two minutes: add the hosted server, then sign in.
Does Cursor wait while Claude Opus 5 runs?
Claude Opus 5 runs in the background. Cursor starts the run, checks on it and gives you the text, json when it's ready.
How much text can Claude Opus 5 handle?
Its context window holds up to 1 million tokens, and its standard maximum output is 128,000 tokens. These are token limits, so avoid treating them as fixed word or page counts.
Which reasoning effort should I use?
Start with high. Test xhigh on difficult coding or extended agent tasks, and compare low or medium on simpler work where your evaluations show acceptable results.
How do I get shorter answers?
Give Claude Opus 5 a concrete length target, such as 150 words or five bullets. Changing reasoning effort does not reliably shorten the visible answer.
How do I keep Claude Opus 5 focused on a small task?
State the exact deliverable and the boundary of the work. For example, ask it to fix one parsing bug while preserving the public interface.
Can Claude Opus 5 give incorrect answers?
Yes. For document analysis, ask for supporting passages and let it say when the supplied material does not establish an answer.
Start using Claude Opus 5 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


