Spiral gives your agent every tool with one account. Browse the tools →

Instinct · LLMReleased 2026-10-07

How to Use Claude Haiku 5.5 in Instinct

Claude Haiku 5.5 suits repetitive summaries, classification, extraction, and focused code edits. For complex coding tasks, Anthropic recommends Claude Sonnet 5.5 or Claude Opus 5.5.

Prompts to try with Claude Haiku 5.5 in Instinct

Copy an example into Instinct, or ask in your own words.

Each morning, use Claude Haiku 5.5 to classify new support messages as billing, access, bug, or other. Report the count and message IDs in each category.

Claude Haiku 5.5Instinct
Try asking
  • Every Friday, use Claude Haiku 5.5 to summarize this week's meeting transcripts into decisions, assigned actions, and unresolved questions.

  • Each evening, use Claude Haiku 5.5 to extract dates and reference numbers from today's incoming messages. Preserve the source message ID for every result.

  • Every hour, use Claude Haiku 5.5 to route new requests to sales, support, or engineering. Flag unclear requests for review.

  • At the end of each day, use Claude Haiku 5.5 to compact today's agent conversations into decisions, constraints, and unfinished work for tomorrow.

How to set up Claude Haiku 5.5 in Instinct

Pick the way that suits you. Each takes about two minutes.

Use the Spiral skill

  1. Send Instinct this message

    Your agent reads Spiral's skill and sets itself up.

    set up https://spiral.ai/SKILL.md
  2. Sign in or paste a key

    Approve Spiral in the browser, or give the agent a key from the console.

Use Spiral's MCP server

  1. Create an API key

    In the console, open API keys and create a key for Instinct. Copy it.

  2. Send Instinct this message

    Instinct reads Spiral's skill and asks for your API key.

    set up https://spiral.ai/SKILL.md
  3. Give Instinct the key

    When Instinct asks for the key, add it where Instinct keeps secrets, never in the chat itself.

Use the Spiral CLI

  1. 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

  1. Create an API key

    In the console, under API keys.

  2. Run Claude Haiku 5.5

    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-haiku-5.5",
        "input": {
          "messages": [
            {
              "role": "system"
            }
          ]
        }
      }'

Then paste one of the prompts above: Instinct finds Claude Haiku 5.5 on Spiral, shows the price and runs it.

Claude Haiku 5.5 providers and prices

Spiral runs Claude Haiku 5.5 through Anthropic. Open a card for today's price, with every fee included.

AnthropicLLM
Claude Haiku 5.5
anthropic/claude-haiku-5.5

Settings you can ask for

Ask Instinct for any of these in your prompt.

SettingAllowed valuesDefaultDescription
Max tokens1 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 effortlow, medium, high, xhigh, or maxmedium

How much the model reasons.

Parallel tool callson or off

Every API field of Claude Haiku 5.5 is on its tool page: open its card above.

What can you do with Claude Haiku 5.5 in Instinct?

Claude Haiku 5.5 can process recurring batches of text for classification, extraction, and summaries. Define the categories and the contents of each recurring report.

Support ticket classification

Assign incoming tickets to billing, access, bug, or other queues.

Meeting summaries

Condense meeting transcripts into decisions, owners, and open questions.

Record extraction

Pull dates, names, and reference numbers from supplied text.

Request routing

Choose the appropriate workflow for each incoming request.

Conversation compaction

Shorten conversation history while preserving decisions and unfinished work.

Focused code edits

Apply a specific change, then run a relevant check.

Claude Haiku 5.5 success rate and speed on Spiral

Success and speed over the last 24 hours, across every run of these tools on Spiral.

Success rate
100.0%
Fastest median run time
3.1s
Runs, last 24 hours
550
ToolSuccessMedian timeRuns
Claude Haiku 5.5100.0%3.1s550

Questions about Claude Haiku 5.5 in Instinct

Can Instinct use Claude Haiku 5.5?

Yes. Add Spiral to Instinct as an MCP server, and Instinct can find and run Claude Haiku 5.5 like any built-in tool.

How much does Claude Haiku 5.5 cost in Instinct?

Up to $0.002672 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, $0.26 per 1M 1-hour 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 Instinct plan is separate.

Do I need an Anthropic account to use Claude Haiku 5.5 in Instinct?

You need no Anthropic account. Instinct uses one Spiral API key for every tool, and Spiral runs Claude Haiku 5.5 on its own Anthropic account.

How long does it take to set up Claude Haiku 5.5 in Instinct?

About two minutes: create an API key, then send Instinct this message, then give Instinct the key.

Does Instinct wait while Claude Haiku 5.5 runs?

Claude Haiku 5.5 runs in the background. Instinct starts the run, checks on it and gives you the text, json when it's ready.

What is Claude Haiku 5.5 best for?

Use it for repeated, clearly defined work: summaries, classification, extraction, routing, and focused code edits.

When should I choose another Claude model?

Anthropic recommends Claude Sonnet 5.5 or Claude Opus 5.5 for complex agentic coding. Keep Haiku tasks narrowly scoped.

How much text can Claude Haiku 5.5 handle?

It has a 1 million token context window and supports up to 128,000 output tokens. Those are token limits, not word counts.

Which reasoning effort should I use?

Medium is the default. High suits longer tasks and work where following detailed instructions matters.

Why can a response end without an answer?

Thinking consumes the output token budget. A small limit can stop generation before visible answer text, so leave room for both thinking and the response.

Get started

Start using Claude Haiku 5.5 in Instinct

One Spiral account for every tool your agent needs. No subscription; you pay per run.

Paste this into Instinct:

Set up Instinct

set up https://spiral.ai/SKILL.md
Are you an AI agent? See setup options