> ## Documentation Index
> Fetch the complete documentation index at: https://docs.extend.ai/llms.txt
> Use this file to discover all available pages before exploring further.
>
> ## API version
> The current API version is `2026-02-09`, served at the site root (no version prefix in URLs).
> If this page URL contains `/2025-04-21/` or `/2024-12-23/`, you are reading an older API version.
> Prefer the current docs at https://docs.extend.ai/llms.txt unless the user explicitly needs that older version.
> Do not treat older-version pages as the source of truth for new integrations.

# Guides

> Ship your document processing with Extend. Get started with our APIs, SDKs, and guides.

## Docs

- [Review Agent](https://docs.extend.ai/extraction/review-agent.md): The Review Agent analyzes extraction results to flag issues and produce an intelligent confidence score that improves on logprobs and OCR-based scores.
- [Operator](https://docs.extend.ai/extraction/operator.md): Agent-guided extraction for long, complex, table-heavy documents.
- [How Credits Work](https://docs.extend.ai/general/how-credits-work.md): How Extend's credit-based billing works: what parse, extract, classify, split, and edit runs cost, how Performance vs Light modes change price, and plan rates.
- [Operator Pricing](https://docs.extend.ai/general/operator-pricing.md): How Operator is billed: a flat page rate plus a passthrough charge for the agent's model usage.
- [Rate Limits](https://docs.extend.ai/general/rate-limits.md): Extend API rate limits by plan tier — per-endpoint request limits, processor throughput, workspace-level overrides, and how to handle 429 responses.
- [User Roles and Permissions](https://docs.extend.ai/general/user-roles-and-permissions.md): Learn about the different user roles in Extend and what each role can access
- [Test Environment Guide](https://docs.extend.ai/general/test-environment-guide.md): Use a test-mode API key to run extractions, classifications, and workflows against an isolated test environment with no production data or webhooks. Covers how test keys route requests, what is isolated vs shared, and how to configure each SDK (Python, TypeScript, Java, Go) for test mode.
- [Reviewing a Workflow Run](https://docs.extend.ai/workflows/reviewing-workflow-run.md): How to review a workflow run once a document has been sent to review
- [Validation Formulas](https://docs.extend.ai/workflows/formulas.md): How to write formulas that reference variables and perform calculations
- [External Data Validation Step](https://docs.extend.ai/workflows/workflow-steps/external-data-validation-step.md): How to leverage the External Data Validation step in a workflow.
- [Conditional Extraction Step](https://docs.extend.ai/workflows/workflow-steps/conditional-extraction-step.md): How to use the Conditional Extraction step to dynamically route documents to different extractors.
- [Evaluation Overview](https://docs.extend.ai/evaluation/overview.md): Evaluation sets measure the accuracy of an extractor, classifier, or splitter against reviewed ground truth. Create sets and run them via the SDK (evaluationSets, evaluationSetItems, evaluationSetRuns) or in Extend Studio.
- [Calculating Array Accuracy](https://docs.extend.ai/evaluation/calculating-array-accuracy.md): How Extend calculates accuracy for array data
- [Composer](https://docs.extend.ai/optimization/composer.md): Composer is an AI assistant that improves your Extend processor configurations — extraction schemas, classifiers, splitters, and editors — from eval feedback.
- [Processor Memory](https://docs.extend.ai/optimization/memory.md): Memory lets Extend classifiers, splitters, and extractors reuse validated historical results to improve accuracy on new documents over time.
- [Webhook Best Practices](https://docs.extend.ai/webhooks/best-practices.md): Operational best practices for handling Extend webhooks