How to Use Claude Fable 5.2 in Amp
Claude Fable 5.2 is built for difficult coding, deep research, and extended technical tasks. Compared with Claude Fable 5.1, Anthropic reports improvements in coding, reasoning, and vision.
Prompts to try with Claude Fable 5.2 in Amp
Copy an example into Amp, or ask in your own words.
Use Claude Fable 5.2 to investigate the failing tests in this repository at high reasoning effort. Reproduce the failure, trace its cause, make a focused fix, and run the relevant checks. Finish with the cause, changes, results, and any remaining blocker.
Use Claude Fable 5.2 to review this project's background job retry logic. Identify paths that could repeat a completed operation, explain the evidence, and propose the smallest correction. Keep the final report under 600 words.
Use Claude Fable 5.2 to plan the database migration described in this project's issue notes. Inspect the schema and callers, identify compatibility risks, and produce an ordered implementation plan with rollback criteria and acceptance checks.
Use Claude Fable 5.2 to compare the two caching approaches documented in this repository. Evaluate invalidation behavior, concurrency, and failure recovery against the project's requirements. Recommend one approach and identify assumptions that need testing.
Use Claude Fable 5.2 to complete the refactor described in the project task. Preserve documented behavior, verify the affected paths, and continue through recoverable errors until the acceptance checks pass or a concrete blocker remains. Report the actual diff and check results.
How to set up Claude Fable 5.2 in Amp
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the server
No header and no client ID or secret: Amp registers itself with Spiral.
amp mcp add spiral https://api.spiral.ai/mcp
- Sign in
Start amp. Amp opens Spiral's sign-in page in your browser: sign in, pick the organization and click Authorize. No API key.
- Check it
amp mcp doctor
Use the Spiral skill
- Send Amp 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 Fable 5.2
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-5.2", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Amp finds Claude Fable 5.2 on Spiral, shows the price and runs it.
Claude Fable 5.2 providers and prices
Spiral runs Claude Fable 5.2 through Anthropic. Open a card for today's price, with every fee included.
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.
Settings you can ask for
Ask Amp 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 Fable 5.2 is on its tool page: open its card above.
What can you do with Claude Fable 5.2 in Amp?
Turn difficult code investigations into reviewed changes with Claude Fable 5.2. Give the agent a defined scope, acceptance checks, and a final report that distinguishes completed work from unresolved issues.
Diagnose a difficult software bug
Ask Claude Fable 5.2 to trace a failure through the relevant code, identify a reproducible cause, and propose a focused fix with verification.
Plan a complex code change
Have Claude Fable 5.2 map dependencies, identify behavior that must stay consistent, and develop an implementation plan with acceptance checks.
Compare technical approaches
Give Claude Fable 5.2 competing designs and explicit requirements. Request a recommendation that identifies assumptions, tradeoffs, and unanswered questions.
Synthesize technical research
Provide source excerpts and ask Claude Fable 5.2 to reconcile competing findings, distinguish evidence from inference, and identify the next questions to investigate.
Turn analysis into a technical report
Ask Claude Fable 5.2 to develop supplied findings into a finished report with recommendations, supporting evidence, and clearly stated gaps.
Review a multistep technical investigation
Have Claude Fable 5.2 assess completed experiments, identify unresolved hypotheses, and produce a prioritized plan for the remaining investigation.
More on Claude Fable 5.2 and Amp
Overview
Claude Fable 5.2 in other agents
Questions about Claude Fable 5.2 in Amp
Can Amp use Claude Fable 5.2?
Yes. Add Spiral to Amp as an MCP server, and Amp can find and run Claude Fable 5.2 like any built-in tool.
How much does Claude Fable 5.2 cost in Amp?
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 Amp plan is separate.
Do I need an Anthropic account to use Claude Fable 5.2 in Amp?
No. Spiral runs Claude Fable 5.2 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 Fable 5.2 in Amp?
About two minutes: add the server, then sign in, then check it.
Does Amp wait while Claude Fable 5.2 runs?
Claude Fable 5.2 runs in the background. Amp starts the run, checks on it and gives you the text, json when it's ready.
What is Claude Fable 5.2 best suited for?
Use it for difficult coding, multistep research, and technical analysis that needs a finished deliverable. Give it a concrete result to produce and clear criteria for completion.
How much context and output can Claude Fable 5.2 handle?
It supports a one-million-token context window and up to 128,000 output tokens. These are token limits, so avoid treating them as fixed page or word counts.
Which reasoning effort should I use?
Start with high effort and compare results on representative tasks. Increase effort for difficult work only when testing shows a benefit; lower it when the results still meet your requirements.
Can I turn Claude Fable 5.2 thinking off?
Adaptive thinking is always enabled. Adjust effort to control thinking depth, and request a specific answer length separately.
What should I do if Claude Fable 5.2 stops before finishing?
State the remaining deliverable and tell it to continue through ordinary, recoverable obstacles. Early stopping is more common at lower effort levels, so test high effort for extended tasks.
Start using Claude Fable 5.2 in Amp
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Amp:
set up https://spiral.ai/SKILL.md


