Exa's 232-line llms.txt shows what thorough AI preparation looks like

# Exa

## Docs

- [Score Deprecation in Auto Search](https://exa.ai/docs/changelog/auto-keyword-score-deprecation.md): We're deprecating relevance scores in Auto search due to architectural improvements. Scores will remain available in Neural search.
- [Auto search as Default](https://exa.ai/docs/changelog/auto-search-as-default.md): Auto search, which intelligently combines Exa's proprietary neural search with other search methods, is now the default search type for all queries.
- [Introducing Exa Company Search](https://exa.ai/docs/changelog/company-search-launch.md): We've added significant improvements to company search due to a fine-tuned retrieval model and entity-matching pipeline. Use `type = "auto"`, `category = "company"` to use this in our search API.
- [Contents Endpoint Status Changes](https://exa.ai/docs/changelog/contents-endpoint-status-changes.md): The /contents endpoint now returns detailed status information for each URL instead of HTTP error codes, providing better visibility into individual content fetch results.
- [Domain Path Filter Support](https://exa.ai/docs/changelog/domain-path-filter.md): `includeDomains` and `excludeDomains` now support URL path filtering and subdomain wildcards.
- [Geolocation Filter Support](https://exa.ai/docs/changelog/geolocation-filter-support.md): `userLocation` added to the search API to bias search results based on geographic location.
- [JS SDK: highlights restored](https://exa.ai/docs/changelog/highlights-restored-js-sdk.md): The highlights feature has been reintroduced in the JavaScript SDK (exa-js) as of version 2.0.11.
- [Added Language Filtering](https://exa.ai/docs/changelog/language-filtering-default.md): Language filtering is now turned on for everyone by default. Exa now detects your query language and only searches web search results in the same language.
- [New Livecrawl Option: Preferred](https://exa.ai/docs/changelog/livecrawl-preferred-option.md): Introducing the 'preferred' livecrawl option that tries to fetch fresh content but gracefully falls back to cached results when crawling fails, providing the best of both worlds.
- [Markdown Contents as Default](https://exa.ai/docs/changelog/markdown-contents-as-default.md): Markdown content is now the default format for all Exa API endpoints, providing cleaner, more readable content that's ideal for AI applications and text processing.
- [New Deep Search Type](https://exa.ai/docs/changelog/new-deep-search-type.md): Introducing Exa Deep: Get better results with smart query expansion and high-quality summaries.
- [New Fast Search Type](https://exa.ai/docs/changelog/new-fast-search-type.md): Introducing Exa Fast: The world's fastest search API.
- [Introducing Exa People Search](https://exa.ai/docs/changelog/people-search-launch.md): We're launching state-of-the-art people search with 1B+ indexed profiles. The 'linkedin' category is now replaced with 'people' for better results.
- [SDK changes: highlights removed and contents returned by default](https://exa.ai/docs/changelog/sdk-major-version-changes.md): Major SDK update with contents included by default in search, highlights feature removed from SDKs, and use_autoprompt field deprecated in all API responses.
- [Company Analyst](https://exa.ai/docs/examples/company-analyst.md): Example project using the Exa Python SDK.
- [Chat app](https://exa.ai/docs/examples/demo-chat.md)
- [Company researcher](https://exa.ai/docs/examples/demo-company-researcher.md)
- [Writing Assistant](https://exa.ai/docs/examples/demo-exa-powered-writing-assistant.md)
- [Hallucination Detector](https://exa.ai/docs/examples/demo-hallucination-detector.md): A live demo that detects hallucinations in content using Exa's search.
- [Websets News Monitor](https://exa.ai/docs/examples/demo-websets-news-monitor.md): A live demo that monitors the web semantically using the Websets API.
- [RAG Q&A](https://exa.ai/docs/examples/exa-rag.md): Using Exa to enable retrieval-augmented generation.
- [Recruiting Agent](https://exa.ai/docs/examples/exa-recruiting-agent.md)
- [Exa Researcher - JavaScript](https://exa.ai/docs/examples/exa-researcher.md): Example project using the Exa JS SDK.
- [Exa Researcher - Python](https://exa.ai/docs/examples/exa-researcher-python.md)
- [Structured Outputs with Instructor](https://exa.ai/docs/examples/getting-started-with-exa-in-instructor.md): Using Exa with instructor to generate structured outputs from web content.
- [Build a Retrieval Agent with LangGraph](https://exa.ai/docs/examples/getting-started-with-rag-in-langgraph.md)
- [Building a Hallucination Checker](https://exa.ai/docs/examples/identifying-hallucinations-with-exa.md): Learn how to build an AI-powered system that identifies and verifies claims using Exa and LangGraph.
- [Job Search with Exa](https://exa.ai/docs/examples/job-search-with-exa.md): Tutorial for simple Exa searches on our front-end.
- [Hacker News Clone](https://exa.ai/docs/examples/live-demo-hacker-news-clone.md): Make your very own Hacker News powered by Exa
- [Phrase Filters: Niche Company Finder](https://exa.ai/docs/examples/niche-company-finder-with-phrase-filters.md)
- [Building a News Summarizer](https://exa.ai/docs/examples/recent-news-summarizer.md): Learn how to build an AI-powered news summarizer that searches and summarizes recent articles using Exa and GPT.
- [Home](https://exa.ai/docs/home.md)
- [Exa](https://exa.ai/docs/integrations/agentops.md)
- [CrewAI Docs](https://exa.ai/docs/integrations/crew-ai-docs.md)
- [Google ADK](https://exa.ai/docs/integrations/google-adk.md)
- [LangChain Docs](https://exa.ai/docs/integrations/langchain-docs.md)
- [LlamaIndex Docs](https://exa.ai/docs/integrations/llamaIndex-docs.md)
- [OpenRouter](https://exa.ai/docs/integrations/openrouter.md)
- [Home](https://exa.ai/docs/introduction.md)
- [Answer](https://exa.ai/docs/reference/answer.md): Get an LLM answer to a question informed by Exa search results. `/answer` performs an Exa search and uses an LLM to generate either:
1. A direct answer for specific queries. (i.e. "What is the capital of France?" would return "Paris")
2. A detailed summary with citations for open-ended queries (i.e. "What is the state of ai in healthcare?" would return a summary with citations to relevant sources)

The response includes both the generated answer and the sources used to create it. The endpoint also supports streaming (as `stream=True`), which will return tokens as they are generated.

Alternatively, you can use the OpenAI compatible [chat completions interface](https://docs.exa.ai/reference/chat-completions#answer).

- [Anthropic Tool Calling](https://exa.ai/docs/reference/anthropic-tool-calling.md): Using Claude's Tool Use Feature with Exa Search Integration.
- [Best Practices](https://exa.ai/docs/reference/best-practices.md): Tips and recommendations for getting the most out of Exa
- [Company Research Claude Skill](https://exa.ai/docs/reference/company-research-claude-skill.md): This guide shows you how to set up a Claude skill and Exa MCP that helps you research companies.
- [Contents Best Practices](https://exa.ai/docs/reference/contents-best-practices.md): Best practices for using Exa's Contents API
- [Get started with Exa](https://exa.ai/docs/reference/contents-quickstart.md): Make your first request to Exa's contents API
- [Contents Retrieval](https://exa.ai/docs/reference/contents-retrieval.md)
- [Context (Exa Code)](https://exa.ai/docs/reference/context.md): Get relevant code snippets and examples from open source libraries and repositories. Search through code repositories to find contextual examples that help developers understand how specific libraries, frameworks, or programming concepts are implemented in practice.
- [Crawling Subpages](https://exa.ai/docs/reference/crawling-subpages.md)
- [CrewAI](https://exa.ai/docs/reference/crewai.md): Learn how to add Exa retrieval capabilities to your CrewAI agents.
- [Error Codes](https://exa.ai/docs/reference/error-codes.md): Reference for common error codes used by the Exa API
- [How to Evaluate Exa Search](https://exa.ai/docs/reference/evaluating-exa-search.md): Comprehensive guide to benchmarking Exa's search API: methodology, optimal settings, datasets, and quality-latency tradeoffs
- [Exa for Google Sheets](https://exa.ai/docs/reference/exa-for-sheets.md)
- [Exa MCP](https://exa.ai/docs/reference/exa-mcp.md)
- [Exa Research](https://exa.ai/docs/reference/exa-research.md): Automate in-depth web research with structured output support.
- [Exa's Capabilities Explained](https://exa.ai/docs/reference/exas-capabilities-explained.md): This page explains some of the available feature functionalities of Exa and some unique ways you might use Exa for your use-case
- [FAQs](https://exa.ai/docs/reference/faqs.md)
- [Find similar links](https://exa.ai/docs/reference/find-similar-links.md): Find similar links to the link provided and optionally return the contents of the pages.
- [Contents](https://exa.ai/docs/reference/get-contents.md): Get the full page contents, summaries, and metadata for a list of URLs.

Returns instant results from our cache, with automatic live crawling as fallback for uncached pages.
- [Welcome to Exa](https://exa.ai/docs/reference/getting-started.md): Exa is a search engine made for AIs.
- [IBM WatsonX](https://exa.ai/docs/reference/ibm-watsonx.md)
- [LangChain](https://exa.ai/docs/reference/langchain.md): How to use Exa's integration with LangChain to perform RAG.
- [Content Freshness](https://exa.ai/docs/reference/livecrawling-contents.md)
- [LlamaIndex](https://exa.ai/docs/reference/llamaindex.md): A quick-start guide on how to add Exa retrieval to a LlamaIndex Agent Application.
- [Migrating from Bing](https://exa.ai/docs/reference/migrating-from-bing.md): Guide for switching from the deprecated Bing Search API to Exa
- [OpenAI Exa Wrapper](https://exa.ai/docs/reference/openai.md): Enhance your OpenAI chat completetions with a simple Exa wrapper that handles search, chunking and prompting.
- [OpenAI Responses API](https://exa.ai/docs/reference/openai-responses-api-with-exa.md): Use Exa with OpenAI's Responses API - both as a web search tool and for direct research capabilities.
- [OpenAI SDK Compatibility](https://exa.ai/docs/reference/openai-sdk.md): Use Exa's endpoints as a drop-in replacement for OpenAI - supporting both chat completions and responses APIs.
- [OpenAI Tool Calling](https://exa.ai/docs/reference/openai-tool-calling.md): Learn to use OpenAI's tool call feature with Exa's Search Integration
- [OpenAPI Specification](https://exa.ai/docs/reference/openapi-spec.md)
- [Get started with Exa](https://exa.ai/docs/reference/quickstart.md): Make your first request to one of Exa's API endpoints
- [Rate Limits](https://exa.ai/docs/reference/rate-limits.md): Default rate limits for Exa API endpoints
- [Create a task](https://exa.ai/docs/reference/research/create-a-task.md): Create an asynchronous research task that explores the web, gathers sources, synthesizes findings, and returns results with citations. Can be used to generate:
1. Structured JSON matching an `outputSchema` you provide.
2. A detailed markdown report when no schema is provided.

The API responds immediately with a `researchId` for polling completion status. For more details, see [Exa Research](/reference/exa-research).

Alternatively, you can use the OpenAI compatible [chat completions interface](/reference/chat-completions#research).

- [Get a task](https://exa.ai/docs/reference/research/get-a-task.md): Retrieve the status and results of a previously created research task.

Use the unique `researchId` returned from `POST /research/v1` to poll until the task is finished.

- [List tasks](https://exa.ai/docs/reference/research/list-tasks.md): Retrieve a paginated list of your research tasks.

The response follows a cursor-based pagination pattern. Pass the `limit` parameter to control page size (max 50) and use the `cursor` token returned in the response to fetch subsequent pages.