Exa

KnowledgePublisher: Exa Labs

Route Exa MCP traffic through Formal to control and log what AI agents do in Exa:

Block tools you don’t want agents to call, such as deep_researcher_start.
Mask sensitive data in tool results, such as emails and phone numbers in web_search_exa responses.
Keep credentials at the Connector, so users and agents never see Exa tokens.
Log every tool call with the user, the MCP client and its parameters.

0. Prepare your Connector

The Connector needs an MCP listener and a hostname with wildcard DNS and TLS. Set them up once by following Deploy the Connector. Every MCP server reuses them.

1. Add Exa MCP server to your Resources

UI
Terraform
Pulumi
API

Open the Admin Catalog and click Add MCP.
Select the Exa template.
Enter a DNS-label Name, such as exa.
Choose the Space containing your Connector, such as default.
Configure upstream authentication as described below, then save.

If no Connector is in the selected Space, the Configuration tab shows MCP clients can’t connect directly yet. The reason reads: No Connector is in this server's Space.

Set space_id to the ID of the Space containing your Connector.

resource "formal_resource" "exa" {
  name                    = "exa"
  technology              = "mcp"
  hostname                = "mcp.exa.ai"
  space_id                = "<SPACE_ID>"
  native_users_v3_enabled = true
  port                    = 443
  tags = {
    "mcp-endpoint" = "/mcp"
  }
}

resource "formal_resource_oauth_settings" "exa" {
  resource_id = formal_resource.exa.id
}

Set spaceId to the ID of the Space containing your Connector.

import * as formal from "@formalco/pulumi";

const resource = new formal.Resource("mcp", {
  name: "exa",
  technology: "mcp",
  hostname: "mcp.exa.ai",
  port: 443,
  spaceId: "<SPACE_ID>",
  tags: { "mcp-endpoint": "/mcp" },
  nativeUsersV3Enabled: true,
});

new formal.ResourceOauthSettings("mcpOAuth", {
  resourceId: resource.id,
});

Create a Resource with the MCP endpoint and its Space ID.

export FORMAL_API_KEY="<YOUR_API_KEY>"
RESOURCE_RESPONSE=$(curl --fail --silent --show-error \
  --request POST \
  https://api.formal.ai/core.v1.ResourceService/CreateResource \
  --header "x-api-key: ${FORMAL_API_KEY}" \
  --header "Content-Type: application/json" \
  --data-binary @- <<JSON
  {
    "name": "exa",
    "hostname": "mcp.exa.ai",
    "port": 443,
    "technology": "mcp",
    "spaceId": "<SPACE_ID>",
    "tags": [
      {
        "key": "mcp-endpoint",
        "value": "/mcp"
      }
    ]
  }
JSON
)
RESOURCE_ID=$(printf '%s' "${RESOURCE_RESPONSE}" | jq -r '.resource.id')
curl --fail --silent --show-error \
  --request POST \
  https://api.formal.ai/core.v1.ResourceService/CreateResourceOAuthSettings \
  --header "x-api-key: ${FORMAL_API_KEY}" \
  --header "Content-Type: application/json" \
  --data-binary @- <<JSON
  {
    "resourceId": "${RESOURCE_ID}"
  }
JSON

2. Setup Authentication

Linked Accounts (Recommended)
Shared Credentials
Pass Through

Each user links their own Exa account once in the Employee Catalog, and the Connector sends that user’s token to Exa.

Linked Accounts selected under How does Formal authenticate to this resource?

Under OAuth app, keep Formal’s App.

Formal’s App (default)

Exa supports Client ID Metadata Documents, so you register nothing. Users see Formal on Exa’s consent screen.

An administrator stores one Exa token in Formal, and the Connector sends it on every user’s requests without showing it to them.

Shared Credentials selected under How does Formal authenticate to this resource?

Exa’s MCP endpoint reads an API key from the x-api-key header. Create one at dashboard.exa.ai/api-keys.

Open this server from the Admin Catalog.
Select Configuration, then Authentication.
Choose Shared Credentials, then select Add credential.
Enter the credential, make it the default, and select Save authentication.

See Native Users.Choose API Key (Header) and set Key Location to Header. Enter x-api-key under Key Name and the raw key under Key Value.In the Terraform tab, set native_users_v3_enabled = true on the Resource and upstream_oauth_mode = "none" on its OAuth settings.

variable "mcp_shared_credential" {
  type      = string
  sensitive = true
}

resource "formal_native_user_v3" "mcp_shared_credential" {
  resource_id = formal_resource.exa.id
  label       = "shared-credential"

  http_api_key_header {
    key = "x-api-key"
    value {
      literal_wo         = var.mcp_shared_credential
      literal_wo_version = 1
    }
  }
}

resource "formal_resource_native_user_selection" "mcp_shared_credential" {
  resource_id = formal_resource.exa.id
  cel         = "\"${formal_native_user_v3.mcp_shared_credential.id}\""
}

3. Route Endpoint traffic

Auth is managed at the Connector level when you use Linked Accounts or Shared Credentials, so the Endpoint needs to forward the MCP traffic to it. Laptops and VMs with the Formal Endpoint keep using https://mcp.exa.ai/mcp, and a network rule sends that traffic to the Connector.

UI
Terraform
Pulumi
API

Go to Network Rules and create a rule named exa-mcp.
Paste this expression, set the rule to Active, then save:

{
  "condition": {
    "pre_tcp": process.is_agent,
    // Every MCP server added to Formal: process.is_agent && resource.technology == "mcp"
    "pre_tls": process.is_agent && hostname == "mcp.exa.ai",
    "post_tls": true
  },
  "outputs": {
    "forward_to_connector": true
  }
}

resource "formal_network_rule" "exa_mcp" {
  name        = "exa-mcp"
  description = "Forward Exa MCP traffic from agents to the Connector"
  status      = "active"

  # Every MCP server added to Formal: set pre_tls to process.is_agent && resource.technology == "mcp"
  cel_expression = <<-EOT
    {
      "condition": {
        "pre_tcp": process.is_agent,
        "pre_tls": process.is_agent && hostname == "mcp.exa.ai",
        "post_tls": true
      },
      "outputs": {
        "forward_to_connector": true
      }
    }
  EOT
}

import * as formal from "@formalco/pulumi";

new formal.NetworkRule("exaMcp", {
  name: "exa-mcp",
  description: "Forward Exa MCP traffic from agents to the Connector",
  status: "active",
  // Every MCP server added to Formal: set pre_tls to process.is_agent && resource.technology == "mcp"
  celExpression: `
    {
      "condition": {
        "pre_tcp": process.is_agent,
        "pre_tls": process.is_agent && hostname == "mcp.exa.ai",
        "post_tls": true
      },
      "outputs": {
        "forward_to_connector": true
      }
    }
  `,
});

export FORMAL_API_KEY="<YOUR_API_KEY>"
# Every MCP server added to Formal: set pre_tls to process.is_agent && resource.technology == "mcp"
CEL_EXPRESSION=$(cat <<'CEL'
{
  "condition": {
    "pre_tcp": process.is_agent,
    "pre_tls": process.is_agent && hostname == "mcp.exa.ai",
    "post_tls": true
  },
  "outputs": {
    "forward_to_connector": true
  }
}
CEL
)
jq -n --arg cel "${CEL_EXPRESSION}" '{
  name: "exa-mcp",
  description: "Forward Exa MCP traffic from agents to the Connector",
  status: "active",
  celExpression: $cel
}' | curl --fail --silent --show-error \
  --request POST \
  https://api.formal.ai/core.v1.DesktopService/CreateDesktopRoutingRule \
  --header "x-api-key: ${FORMAL_API_KEY}" \
  --header "Content-Type: application/json" \
  --data-binary @-

The rule matches known agents only. See Identity headers.Cloud agents such as Claude, ChatGPT and Cursor Cloud Agents don’t run the Endpoint. Point them at this server’s Connector address, https://exa.<connector-hostname>/mcp. Find it in the Admin Catalog: open Exa, then copy Connector address on the Configuration tab.

4. Create policies

Tools

The Admin Catalog lists these tools for Exa:

Tool	Description
`web_search_exa`	Search the web with Exa.
`get_code_context_exa`	Find code examples and technical context.
`research_paper_search`	Search research papers.
`company_research`	Research a company.
`crawling`	Retrieve clean content from a URL.
`deep_researcher_start`	Start a multi-step research task.
`deep_researcher_check`	Check a research task and retrieve results.

5. Check the logs

Open Logs and search for resource.name:exa. Each Exa tool call is one request log with the user, the MCP client and the tool. Open a log to see its tool parameters and the policies it triggered.See Logs for search syntax and filters.

Troubleshooting

The agent says Exa rejected your credentials

The agent relays: exa rejected your credentials. Link or re-link your exa account in Formal…Cause: Exa no longer accepts the linked token, for example after the user revoked access in Exa.Fix: Link Exa again from the Employee Catalog.With Shared Credentials, the message asks the user to contact an administrator. Replace the token in the server’s Shared Credentials.

Laptop traffic doesn't appear in Logs

Cause: The Endpoint isn’t running, or the network rule doesn’t match the client.Fix: Check the Endpoint is connected. The rule above only matches known agents. Other clients, such as a browser, reach Exa directly unless the rule also matches them. See Network rules.

MCP clients can't connect directly yet

Cause: The Configuration tab lists the reason, most often No Connector is in this server's Space.Fix: Edit the server and choose the Space containing its Connector.

Getting Started

Example Use Cases

AI Governance

Configuring Formal

Integrations

Configuration

0. Prepare your Connector

1. Add Exa MCP server to your Resources

2. Setup Authentication

Formal’s App (default)

3. Route Endpoint traffic

4. Create policies

Tools

5. Check the logs

Troubleshooting

Next Steps

MCP Gateway

All MCP servers

​0. Prepare your Connector

​1. Add Exa MCP server to your Resources

​2. Setup Authentication

​Formal’s App (default)

​3. Route Endpoint traffic

​4. Create policies

​Tools

​5. Check the logs

​Troubleshooting

​Next Steps

MCP Gateway

All MCP servers

0. Prepare your Connector

1. Add Exa MCP server to your Resources

2. Setup Authentication

Formal’s App (default)

3. Route Endpoint traffic

4. Create policies

Tools

5. Check the logs

Troubleshooting

Next Steps