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

Instinct · LLM

How to Use Claude Fable Latest in Instinct

Use Claude Fable Latest for difficult coding, research, and analysis that needs several rounds of checking. Anthropic currently identifies Fable 5.2 as its latest release, with support for large software migrations and mathematical proof formalization.

Prompts to try with Claude Fable Latest in Instinct

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

Use Claude Fable Latest each morning to analyze the experiment summaries supplied for that run. Compare them with the baseline, flag results that need replication, and finish with a prioritized ablation plan.

Claude Fable LatestInstinct
Try asking
  • Use Claude Fable Latest each week to synthesize the technical research excerpts supplied for that run. Produce a 700-word brief covering changed conclusions, conflicting evidence, and unanswered questions.

  • Use Claude Fable Latest with high reasoning effort each night to review the incident timelines supplied for that run. Test competing explanations against the evidence and produce a ranked investigation plan.

  • Use Claude Fable Latest each Friday to review the supplied migration status summaries. Compare completed steps with acceptance criteria and produce a report of verified progress and remaining blockers.

  • Use Claude Fable Latest weekly to review the mathematical arguments supplied for that run. List unresolved assumptions and missing lemmas, then propose the next checkable proof steps.

How to set up Claude Fable Latest 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 Fable Latest

    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-fable-latest",
        "input": {
          "messages": [
            {
              "role": "system"
            }
          ]
        }
      }'

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

Claude Fable Latest providers and prices

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

AnthropicLLM
Claude Fable Latest
~anthropic/claude-fable-latest

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 maxhigh

How much the model reasons.

Parallel tool callson or off

Every API field of Claude Fable Latest is on its tool page: open its card above.

What can you do with Claude Fable Latest in Instinct?

Carry a recurring technical investigation through to a finished report with Claude Fable. Give each run a baseline and completion checklist so it can compare evidence and identify unfinished work.

Plan and implement a software migration

Map existing behavior, migrate one component at a time, and compare results against the original implementation.

Investigate a persistent software bug

Trace a failure across components, test competing explanations, and produce a fix with evidence that it addresses the cause.

Review architecture across repositories

Compare dependencies and responsibilities across services, then turn the findings into a concrete change proposal.

Produce a technical research brief

Work from supplied sources to develop an analysis, resolve conflicting evidence, and finish a document with clear conclusions.

Evaluate machine learning experiments

Compare a baseline with proposed changes, plan ablations, and check whether measured improvements survive repeat runs.

Develop a checkable mathematical proof

Break a proof into lemmas, state assumptions, and work toward a formalization that a proof checker can validate.

Questions about Claude Fable Latest in Instinct

Can Instinct use Claude Fable Latest?

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

How much does Claude Fable Latest cost in Instinct?

Up to $0.21372 per run at the default inputs. $10.40 per 1M input tokens, $0.26 per 1M cached input tokens, $13.00 per 1M cache write tokens, $20.80 per 1M 1-hour cache write tokens, $52.00 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Instinct plan is separate.

Do I need an Anthropic account to use Claude Fable Latest in Instinct?

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

How long does it take to set up Claude Fable Latest in Instinct?

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

Does Instinct wait while Claude Fable Latest runs?

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

What is Claude Fable best suited for?

Difficult software changes, extended research, and analysis that needs repeated checking. Anthropic's current release, Fable 5.2, also supports work across large repositories.

How much context and output can Claude Fable handle?

The current Fable 5.2 release supports a one-million-token context window and up to 128,000 output tokens. Thinking uses part of the output allowance, so the maximum is not a promise of 128,000 tokens of visible text.

Which reasoning effort should I start with?

Start at high and evaluate the results on your actual tasks before increasing it. Leave enough output room for both thinking and the finished answer.

What weakness should I watch for on long tasks?

Fable 5.2 can end a turn while work remains. Give it a completion checklist and explicitly ask it to work through ordinary errors until the requested deliverable is finished.

Can Claude Fable help with mathematical proofs?

Yes. Anthropic reports that a team of Fable 5.2 agents formalized Fermat's Last Theorem; for your own work, ask for explicit assumptions and verification with a proof checker where available.

Get started

Start using Claude Fable Latest 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