The server itself exposes 4 tools, and some of them can write to your Exa account. Behind TrustGate an agent only sees the tools you allow, requests and responses are inspected, and every call is logged. Anything outside the allowlist is denied before it reaches Exa.

How is this different from connecting Exa MCP directly?

Connected directly, an agent sees every tool the server exposes and holds the credential itself. Through TrustGate you choose which tools each agent gets, and every call is logged against the agent that made it.

What if the Exa MCP API changes?

We track the Exa MCP schema, so the tool catalogue stays current. Tools you have already allowed keep working, and anything new stays denied until you enable it.

Can I give different agents different Exa tools?

Yes. Access is set per agent: one can get a handful of read tools, another the full set. You grant the whole MCP or single tools within it.

Where can I see what my agents did in Exa?

Every call is logged against the agent that made it, with the tool it used. Activity can be sent on to your own monitoring.

Do I need to run the Exa MCP server myself?

No. TrustGate connects to the Exa MCP server and sits in front of it. Your agent only talks to your TrustGate URL, never to Exa directly.

Connect Exa MCP to your AI agents, securely.

One secured endpoint for every Exa call your AI agents make, scoped per user, inspected, and logged.

Get started for free

Get a demo

Try Exa MCP now

What your agents can ask once Exa MCP is connected.

Select the tools your AI agents can call

4 tools from the Exa MCP schema.

Showing 4 tools

web_search_exareadSearches the web for any topic and returns clean, ready-to-use content. Enabled by default.
web_fetch_exareadReads a webpage's full content as clean markdown, from one or more URLs. Enabled by default.
web_search_advanced_exareadAdvanced search with filters, domains, dates, highlights, summaries and subpage crawling. Off by default; enable with the tools parameter.
agent_runhigh riskRuns an Exa Agent for multi-step research, list-building, enrichment and structured output. Off by default and requires OAuth or an API key.

What Exa MCP exposes

Exa MCP groups its 4 tools into 2 categories, each with a suggested starting policy.

All 4 tools as JSON, ready to paste.

Category	Covers	Access	Suggested policy
Search & Fetch	Web search returning clean content, advanced search with domain, date, highlight, summary and subpage-crawl filters, and reading one or more pages as markdown	Read	Treat everything returned as untrusted third-party text, never as instructions. Only the two basic tools are on by default
Agent	Running an Exa Agent for multi-step research, list-building, enrichment and structured output	Read / Write	Off by default and the only tool that requires authentication; an agent run is open-ended, so approve and bound it

Connect Exa MCP to your AI agent

Four steps, under ten minutes.

01
Sign in to NeuralTrust
Open the NeuralTrust platform, sign in and go to Getting started, then Connect your tools.
02
Choose your agent
When asked which agent you use, pick Claude Code, Codex, Cursor, Windsurf, Copilot or your own code.
03
Connect Exa
Find Exa in the tool list and connect it.
04
Connect your agent
Follow the steps on screen to add the config to your agent, then send the test prompt.

Permissions down to the single tool

You do not allow Exa MCP or block it. You allow the exact tools each agent can call, one by one.

Exa MCP	4 TOOLS AVAILABLE
Code review agent	4 tools, read only
Issue triage bot	9 tools
Platform team assistant	all 4 tools

Tool by toolConnecting the server is not the unit of access. Each tool is granted on its own.
Default denyA tool you have not enabled is never exposed to the agent.
Approval on writesTools that change or delete things can require a human to confirm.

Is Exa MCP secure?

One endpoint instead of raw credentials. Every call is checked against policy and logged.

AI agentsConnect through your MCP client configuration.	NeuralTrustApply controls based on your configuration.	Exa MCP4 documented tools. Two are enabled by default; the tools URL parameter replaces that default set rather than extending it.
Coding assistants	Connection & credentialsper user	Search & Fetch3 tools
Internal copilots	Tool access policiesdeny by default	Agent1 tool
Workflow automations	Request & response inspectioninline
Compatible MCP clients	Activity monitoringalerts and logs