llms.txt snapshot

# Exa

> Exa is a real-time knowledge index and retrieval platform built for AI agents. Exa builds and maintains its own index and search models, and continuously indexes the live web alongside structured data about companies, people, financial markets, code, documents, and other sources. Developers use Exa’s APIs to search, retrieve, and research public and private data.

**Start for free:** New accounts receive $20 in free credits and $10 in free credits every month. No payment method is required, and the free tier includes access to all API endpoints.

For AI agents and automation, start with the agent-native interfaces below. For complete technical documentation, use the Exa documentation index.

## MCP Server

Connect Exa search, code search, content retrieval, and research tools to Claude, Cursor, Codex, VS Code, and other MCP-compatible clients.

The hosted MCP server can be tried without an API key at limited usage levels.

- [Exa MCP](https://exa.ai/mcp)
- [MCP documentation](https://exa.ai/docs/reference/exa-mcp.md)
- [Hosted MCP server](https://mcp.exa.ai/mcp)
- [MCP server repository](https://github.com/exa-labs/exa-mcp-server)

## Agent Skills

Install reusable skills that teach coding agents how to search, retrieve content, research companies, generate leads, and build applications with Exa.

- [Agent Skills](https://exa.ai/docs/reference/agent-skills.md)
- [Agent Skills repository](https://github.com/exa-labs/agent-skills)

## Documentation

Exa's complete technical documentation, including quickstarts, coding-agent references, API guides, SDKs, integrations, examples, and API reference.

Use `/docs/llms.txt` to discover the full technical documentation corpus. Exa docs pages are also available as Markdown by appending `.md` to the docs URL.

- [Documentation index](https://exa.ai/docs/llms.txt)
- [Full documentation corpus](https://exa.ai/docs/llms-full.txt)
- [API dashboard and keys](https://dashboard.exa.ai/api-keys)

## Products

### Search API

Search the web and extract contents from the results. Built for AI applications and agents. Use Deep Search for multi-step web search with reasoning and structured outputs for complex research tasks.

- [Search API](https://exa.ai/products/search)
- [Search documentation](https://exa.ai/docs/reference/search-api-guide.md)
- [Deep Search](https://exa.ai/products/deep)
- [Search reference for coding agents](https://exa.ai/docs/reference/search-api-guide-for-coding-agents.md)

### Contents API

Extract clean, LLM-ready web content from any URL, including JavaScript-rendered pages, PDFs, and complex layouts.

- [Contents API documentation](https://exa.ai/docs/reference/contents-api-guide.md)

### Agent

Run multi-step web research, list-building, and enrichment workflows with natural-language answers, structured outputs, and citations.

- [Agent](https://exa.ai/products/agent)
- [Agent documentation](https://exa.ai/docs/reference/agent-api-guide.md)

### Exa Connect

Connect your agents to the world's public and private data, including web analytics, private markets, company and contact data, financial data, healthcare intelligence, compliance data, and other specialized datasets from leading providers.

- [Exa Connect](https://exa.ai/connect)
- [Connect documentation](https://exa.ai/docs/reference/agent-api/connect/overview.md)

## Use Cases

Exa powers web search and data retrieval for coding agents, research agents, GTM workflows, chatbots, voice agents, news monitoring, people search, company search, and enrichment.

- [GTM and enrichment](https://exa.ai/use-cases/gtm)
- [Demos](https://exa.ai/demos)

## Integrations

Use Exa with agent frameworks, SDKs, automation tools, AI assistants, and developer platforms including OpenAI, Vercel AI SDK, LangChain, CrewAI, LlamaIndex, Pydantic AI, n8n, and others.

Coding agents can also use Stripe Projects for agentic onboarding to create an Exa account, provision an API key, and sync it directly into a project.

- [Integrations directory](https://exa.ai/integrations)
- [Agentic onboarding with Stripe](https://exa.ai/docs/reference/stripe-projects)

## SDKs

Official Exa SDKs for Python and JavaScript/TypeScript.

- [Python SDK](https://github.com/exa-labs/exa-py)
- [JavaScript SDK](https://github.com/exa-labs/exa-js)
- [Python SDK documentation](https://exa.ai/docs/sdks/python-sdk.md)
- [JavaScript SDK documentation](https://exa.ai/docs/sdks/javascript-sdk.md)

## OpenAPI

Machine-readable API specifications. Use the OpenAPI specification as the source of truth for request and response schemas; use the guides for recommended usage patterns and defaults.

- [OpenAPI specification (JSON)](https://exa.ai/docs/exa-spec.json)
- [OpenAPI specification (YAML)](https://exa.ai/docs/exa-spec.yaml)
- [Team Management API specification](https://exa.ai/docs/team-management-spec.yaml)

## Pricing

Exa can be started for free. New accounts receive $20 in credits at signup and $10 in free credits every month, with no payment method required. The free tier includes access to all endpoints.

Paid API usage is pay-as-you-go, with enterprise plans available for higher limits, custom infrastructure, security requirements, and support.

- [Pricing](https://exa.ai/pricing)
- [Pricing documentation](https://exa.ai/docs/reference/pricing.md)
- [Billing](https://exa.ai/docs/reference/billing.md)
- [Rate limits](https://exa.ai/docs/reference/rate-limits.md)

## Evals and Benchmarks

Exa publishes evaluations of search quality, latency, vertical retrieval, coding-agent search, and agentic research. Where available, datasets and evaluation harnesses are open source and reproducible.

- [Evals](https://exa.ai/evals)
- [Open-source benchmarks](https://github.com/exa-labs/benchmarks)

## Comparisons

Detailed comparisons of Exa with other search, retrieval, and research APIs include product differences, pricing, benchmarks, and recommended use cases.

- [Exa vs Tavily](https://exa.ai/versus/tavily)
- [Exa vs Parallel](https://exa.ai/versus/parallel)
- [Exa vs Perplexity](https://exa.ai/versus/perplexity)
- [Exa vs Firecrawl](https://exa.ai/versus/firecrawl)
- [Exa vs Google Search](https://exa.ai/versus/google-search)
- [Exa vs Brave](https://exa.ai/versus/brave)
- [Exa vs SerpAPI](https://exa.ai/versus/serpapi)
- [Exa vs Serper](https://exa.ai/versus/serper)
- [Exa vs You.com](https://exa.ai/versus/you)

## Customers

See how production AI products and agents use Exa for web search, research, coding, company and people data, and enrichment.

- [Customer stories](https://exa.ai/customers)

## Research

Technical research and engineering work from Exa on web-scale retrieval, search models, evaluation, search infrastructure, and agentic search.

- [Research](https://exa.ai/research)
- [Blog](https://exa.ai/blog)

## Company and Trust

Exa is an applied AI research lab building search infrastructure for AI systems.

- [About Exa](https://exa.ai/about)
- [Security](https://exa.ai/security)
- [Trust Center](https://trust.exa.ai)
- [GitHub](https://github.com/exa-labs)
- [Changelog](https://exa.ai/docs/changelog.md)