Skip to main content The #1 agentic semantic tool search: 91.6% first-try accuracy on S1 Search Bench Explore Tool Discovery
Live 35 Actions

Exa MCP Server
for AI Agents

Connect your AI agent to StackOne's Exa MCP server and give it 35 MCP tools out of the box. Auth, tool execution, and security all managed.

Exa logo
Exa MCP Server
Built by StackOne StackOne
DrataGPLocalyzeFlipMindtoolsScreenloop

Coverage

35 Agent Actions

Create, read, update, and delete across Exa — and extend your agent's capabilities with custom actions.

Authentication

Agent Tool Authentication

Per-user OAuth in one call. Your Exa MCP server gets session-scoped tokens with zero credentials stored on your infra.

Agent Auth →

Security

Agent Protection

Every Exa tool response scanned for prompt injection in milliseconds — 88.7% accuracy, all running on CPU.

Prompt Injection Defense →

Performance

Max Agent Context. Min Cost.

Free up to 96% of your agent's context window to enhance reasoning and reduce cost, on every Exa call.

Tools Discovery →

What is the Exa MCP Server?

A Exa MCP server lets AI agents read and write Exa data through the Model Context Protocol — Anthropic's open standard for connecting LLMs to external tools. StackOne's Exa MCP server ships with 35 pre-built actions, fully extensible via the Connector Builder — plus managed authentication, prompt injection defense, observability, and agent execution runtime. Connect it from MCP clients like Claude Desktop, Claude Code, Cursor, Goose, and VS Code, or from agent frameworks like OpenAI Agents SDK, LangChain, and Vercel AI SDK.

All Exa MCP Tools

Every action from Exa's API, ready for your agent. Create, read, update, and delete — scoped to exactly what you need.

Events

  • List Events

    List all events that have occurred in the system. You can paginate through the results using the `cursor` parameter.

  • Get Event

    Get a single Event by id. You can subscribe to Events by creating a Webhook.

Monitor Runs

  • List Monitor Runs

    Lists all runs for the Monitor.

  • Get Monitor Run

    Gets a specific monitor run.

Monitors

  • Create Monitor

    Create a Monitor that re-runs a Webset search on a schedule to keep the Webset updated with fresh items.

  • List Monitors

    Lists all monitors for the Webset.

  • Get Monitor

    Retrieve a single Websets Monitor by its ID, including its schedule and behavior.

  • Update Monitor

    Updates a monitor configuration.

  • Delete Monitor

    Permanently delete a Websets Monitor by its ID.

Webset Enrichments

  • Create Webset Enrichment

    Create an Enrichment for a Webset.

  • Get Webset Enrichment

    Retrieve a single enrichment on a Webset by the Webset ID and enrichment ID.

  • Update Webset Enrichment

    Update an Enrichment configuration for a Webset.

  • Delete Webset Enrichment

    When deleting an Enrichment, any running enrichments will be canceled and all existing `enrichment_result` generated by this Enrichment will no longer be available.

Webset Items

  • List Webset Items

    Returns a list of Webset Items. You can paginate through the Items using the `cursor` parameter.

  • Get Webset Item

    Retrieve a single item within a Webset by the Webset ID and item ID.

  • Delete Webset Item

    Deletes an Item from the Webset. This will cancel any enrichment process for it.

Webset Searchs

  • Create Webset Search

    Creates a new Search for the Webset. The default behavior is to reuse the previous Search results and evaluate them against the new criteria.

  • Get Webset Search

    Retrieve a single search on a Webset by the Webset ID and search ID.

Websets

  • Create Webset

    Creates a new Webset with optional search, import, and enrichment configurations. The Webset will automatically begin processing once created. You can specify an `externalId` to reference the Webset with your own identifiers for easier integration.

  • List Websets

    Returns a list of Websets. You can paginate through the results using the `cursor` parameter.

  • Get Webset

    Retrieve a single Webset by its ID, optionally expanding its items.

  • Update Webset

    Update the metadata of an existing Webset identified by its ID.

  • Delete Webset

    Deletes a Webset. Once deleted, the Webset and all its Items will no longer be available.

Other (12)

  • Create Import

    Create an Import to upload external data into Websets for enrichment, search, or exclusion.

  • Get Contents

    Retrieve cleaned page contents (text, highlights, and/or summary) for one or more URLs or Exa result IDs.

  • List Imports

    Lists all imports for the Webset.

  • Get Import

    Retrieve a single Websets Import by its ID, including its status and configuration.

  • Search

    Perform a search with an Exa prompt-engineered query and retrieve a list of relevant results. Optionally get contents.

  • Update Import

    Updates a import configuration.

  • Delete Import

    Permanently delete a Websets Import by its ID.

  • Generate Answer

    Performs a search based on the query and generates either a direct answer or a detailed summary with citations, depending on the query type.

  • Cancel Webset Enrichment

    All running enrichments will be canceled. You can not resume an Enrichment after it has been canceled.

  • Cancel Webset Search

    Cancels a currently running Search. You can cancel all searches at once by using the `websets/:webset/cancel` endpoint.

  • Cancel Webset

    Cancels all operations being performed on a Webset. Any enrichment or search will be stopped and the Webset will be marked as `idle`.

  • Preview Webset

    Preview how a search query will be decomposed before creating a webset. This endpoint performs the same query analysis that happens during webset creation, allowing you to see the detected entity type, generated search criteria, and available enrichment columns in advance. Use this to help users understand how their search will be interpreted before committing to a full webset creation.

Set Up Your Exa MCP Server in Minutes

One endpoint. Any framework. Your agent is talking to Exa in under 10 lines of code.

Agent Frameworks

Claude Desktop
{
  "mcpServers": {
    "stackone": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote@latest",
        "https://api.stackone.com/mcp?x-account-id=<account_id>",
        "--header",
        "Authorization: Basic <YOUR_BASE64_TOKEN>"
      ]
    }
  }
}

Check More AI & ML MCP Servers

OpenAI

121+ actions

Dialpad

57+ actions

Google Gemini

30+ actions

Anthropic

29+ actions

Fireflies.ai

25+ actions

Groq

20+ actions

Cohere

14+ actions

Exa MCP Server FAQ

Does StackOne have a Exa MCP server?
Yes. StackOne offers a hosted Exa MCP server with 35 pre-built actions, and every action is tested and QA'd by StackOne. Connect it to Claude, Cursor, and any other MCP client, or to any agent framework through the AI Action SDK. It ships with managed agent authentication, prompt injection defense, and tool discovery with server-side execution that preserve your agent's context window and keep reasoning performance.
Exa MCP server vs direct API integration — what's the difference?
A Exa MCP server and direct API integration serve different use cases. Direct API integration is for software-to-software — backend code calling Exa. A Exa MCP server is for AI agents — MCP clients like Claude and Cursor, plus framework agents built with OpenAI, LangChain, or Vercel AI — discovering and calling Exa at runtime. StackOne provides both.
How does Exa authentication work for AI agents?
Exa authentication for AI agents works through a StackOne Connect Session. Create one via the dashboard or the SDK — you get an auth link and ready-to-paste config for Claude Desktop, Cursor, and other MCP clients. Your user authenticates their own Exa account; StackOne handles token exchange, storage, and refresh. Credentials never reach the LLM, and each user is isolated via origin_owner_id.
Are Exa MCP tools vulnerable to prompt injection?
Yes — Exa MCP tools can be vulnerable to indirect prompt injection. Any tool that reads user-written content — documents, messages, tickets, records, or free-text fields — is a potential vector. StackOne Defender scans every tool response before it enters the agent's context — regex patterns in ~1ms, then a MiniLM classifier in ~4ms. 88.7% accuracy, CPU-only.
What is the context bloat of a Exa agent and how do I avoid it?
Context bloat happens when Exa tool schemas and API responses eat your Exa agent's memory, preventing it from reasoning effectively. A single Exa query can return a massive JSON response, and connecting multiple tools compounds the problem. Tools Discovery and Code Mode reduce context bloat — loading only relevant tools per query and keeping raw responses out of the agent's context.
Can I limit which actions my Exa agent can access?
Yes — you can limit which actions your Exa agent can access directly from the StackOne dashboard. Toggle actions on or off, or restrict them to specific accounts, with no code changes to your agent. Session tokens can be scoped to exact actions so if one leaks, exposure stays contained.
Can I create custom agent actions for my Exa MCP server?
Yes — you can create custom agent actions for your Exa MCP server using Connector Builder. It's an integration agent your coding assistant (Claude Code, Cursor, or Copilot) can invoke to research Exa's API, generate production-ready connector YAML, test against the live API, and validate before you ship.
When should I NOT use a Exa MCP server?
Skip a Exa MCP server if your integration is purely software-to-software — direct Exa API integration is simpler when no AI agent is involved. For deterministic, compliance-critical operations (financial transactions, regulatory reporting), direct API gives you predictable behavior without agent-driven decision-making. MCP shines when AI agents need to dynamically discover and call Exa actions at runtime.
What AI frameworks and AI clients does the StackOne Exa MCP server support?
The StackOne Exa MCP server supports both. MCP clients (paste-and-go apps): Claude Desktop, Claude Code, Cursor, VS Code, Goose. Agent frameworks (code SDKs you build with): OpenAI Agents SDK, Anthropic, Vercel AI, Google ADK, CrewAI, Pydantic AI, LangChain, LangGraph, Azure AI Foundry.

Put your AI agents to work

All the tools you need to build and scale AI agent integrations, with best-in-class connectivity, execution, and security.