How to Use Claude Sonnet 5 in opencode
Claude Sonnet 5 suits coding, debugging, and tasks that require several tool calls. Compared with Claude Sonnet 4.6, it improves reasoning, tool use, coding, and knowledge work.
Prompts to try with Claude Sonnet 5 in opencode
Copy an example into opencode, or ask in your own words.
Use Claude Sonnet 5 at high effort to investigate the failing tests in this project. Reproduce the failure, trace its cause, make the smallest justified fix, and run the relevant checks.
Use Claude Sonnet 5 at extra high effort to inspect this project's background job retry logic. Find paths that could process the same job twice, propose a fix, and verify it with a targeted test.
Use Claude Sonnet 5 to review the current diff for incorrect behavior. For every finding, give the file location, triggering conditions, expected impact, and a test that would expose it.
Use Claude Sonnet 5 at high effort to implement pagination for this project's existing list endpoint. Follow the repository's conventions, preserve existing behavior, and test empty, first, and final pages.
Use Claude Sonnet 5 to compare the requirements in this project's design documents with the implementation. List unmet requirements with document references and code locations; apply the same review criteria to every requirement.
How to set up Claude Sonnet 5 in opencode
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the server
Leave out --global to add Spiral to the current project only. No header and no client ID or secret: opencode registers itself with Spiral.
opencode mcp add spiral --global --url https://api.spiral.ai/mcp
- Sign in
Your browser opens. Sign in to Spiral, pick the organization and click Authorize, then return to the terminal. No API key. Inside opencode, /mcps starts the same sign-in.
opencode mcp auth spiral
- Check it
Spiral shows as connected.
opencode mcp list
Use the Spiral skill
- Send opencode 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 Sonnet 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-sonnet-5", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: opencode finds Claude Sonnet 5 on Spiral, shows the price and runs it.
Claude Sonnet 5 providers and prices
Spiral runs Claude Sonnet 5 through Anthropic. Open a card for today's price, with every fee included.
Up to $0.05343 per run at the default inputs. $2.60 per 1M input tokens, $0.26 per 1M cached input tokens, $3.25 per 1M cache write tokens, $5.20 per 1M 1-hour cache write tokens, $13.00 per 1M output tokens; for a prompt above 200,000 tokens, $5.20 per 1M input tokens, $0.52 per 1M cached input tokens, $6.50 per 1M cache write tokens, $10.40 per 1M 1-hour cache write tokens, $19.50 per 1M output tokens.
Settings you can ask for
Ask opencode 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 Sonnet 5 is on its tool page: open its card above.
What can you do with Claude Sonnet 5 in opencode?
Turn a reported bug or feature request into a code change and a verification report with Claude Sonnet 5. Give it the project constraints and completion criteria so it can plan the work and use the available development tools.
Trace a bug to its cause
Use Claude Sonnet 5 to investigate a failing behavior, propose a fix, and check it against a reproducing test.
Implement a change across a codebase
Give Claude Sonnet 5 the requirements and project constraints, then ask it to update the affected code and verify the result.
Review code for behavioral bugs
Ask Claude Sonnet 5 to examine a change for incorrect behavior, explain each finding, and suggest a test that would expose it.
Compare lengthy specifications
Have Claude Sonnet 5 compare supplied specifications, identify conflicting requirements, and produce a decision table with source references.
Extract records from business text
Ask Claude Sonnet 5 to extract owners, deadlines, and commitments from supplied text, applying the same rules to every record.
Research a technical decision
Use Claude Sonnet 5 with your agent's search tools to gather current documentation and compare implementation choices against explicit requirements.
Prepare recurring operational briefs
Have Claude Sonnet 5 review newly supplied updates, identify unresolved issues, and draft a brief with evidence and proposed next actions.
More on Claude Sonnet 5 and opencode
Claude Sonnet 5 in other agents
Questions about Claude Sonnet 5 in opencode
Can opencode use Claude Sonnet 5?
Yes. Add Spiral to opencode as an MCP server, and opencode can find and run Claude Sonnet 5 like any built-in tool.
How much does Claude Sonnet 5 cost in opencode?
Up to $0.05343 per run at the default inputs. $2.60 per 1M input tokens, $0.26 per 1M cached input tokens, $3.25 per 1M cache write tokens, $5.20 per 1M 1-hour cache write tokens, $13.00 per 1M output tokens; for a prompt above 200,000 tokens, $5.20 per 1M input tokens, $0.52 per 1M cached input tokens, $6.50 per 1M cache write tokens, $10.40 per 1M 1-hour cache write tokens, $19.50 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your opencode plan is separate.
Do I need an Anthropic account to use Claude Sonnet 5 in opencode?
No. Spiral runs Claude Sonnet 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 Sonnet 5 in opencode?
About two minutes: add the server, then sign in, then check it.
Does opencode wait while Claude Sonnet 5 runs?
Claude Sonnet 5 runs in the background. opencode starts the run, checks on it and gives you the text, json when it's ready.
What is Claude Sonnet 5 good at?
Coding, debugging, reasoning, and work that takes several tool calls. Anthropic reports improvements over Claude Sonnet 4.6 in coding, tool use, reasoning, and knowledge work.
How much text can Claude Sonnet 5 handle?
It supports a 1 million token context window and up to 128,000 output tokens per response.
Which effort level should I choose?
Use high effort for complex reasoning and coding, and extra high for the hardest coding or agent tasks. Low effort fits short tasks where response speed matters more than deeper reasoning.
Where can Claude Sonnet 5 fall short?
Low effort can lead to shallow reasoning on moderately complex problems. It also interprets instructions literally, so explicitly say when a rule applies to every item or section.
Does Claude Sonnet 5 know about recent events?
Its reliable knowledge cutoff is January 2026. Supply current source material or ask your agent to retrieve it when an answer depends on newer information.
Start using Claude Sonnet 5 in opencode
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into opencode:
set up https://spiral.ai/SKILL.md


