How is this different from connecting Monte Carlo 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 Monte Carlo MCP API changes?

We track the Monte Carlo 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 Monte Carlo 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 Monte Carlo?

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 Monte Carlo MCP server myself?

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

Connect Monte Carlo MCP to your AI agents, securely.

Q: Is Monte Carlo MCP secure?

The server itself exposes 24 tools, and some of them can write to your Monte Carlo 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 Monte Carlo.

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

Get started for free

Get a demo

Try Monte Carlo MCP now

What your agents can ask once Monte Carlo MCP is connected.

Select the tools your AI agents can call

24 tools from the Monte Carlo MCP schema.

Showing 24 tools

searchreadFinds tables and other data assets by name.
getTablereadGets a table's schema, stats and metadata.
getQueriesForTablereadGets the recent query history that reads from or writes to a table.
getQueryDatareadGets the full SQL and details of a specific query.
getAssetLineagereadGets the upstream or downstream dependencies of one or more assets.
getDownstreamBiReportsreadGets the BI reports and dashboards downstream of an asset.
getMonitorsreadGets monitor configurations, filterable by table.
getAudiencesreadLists the notification audiences that monitors and alerts can be routed to.
createValidationMonitorMacreadGenerates monitors-as-code YAML for a validation monitor.
createMetricMonitorMacreadGenerates monitors-as-code YAML for a metric monitor.
createComparisonMonitorMacreadGenerates monitors-as-code YAML for a comparison monitor.
createCustomSqlMonitorMacreadGenerates monitors-as-code YAML for a custom SQL monitor.
getAlertsreadGets active incidents and alerts.
updateAlertwriteUpdates an alert.
setAlertOwnerwriteAssigns an owner to an alert.
createOrUpdateAlertCommentwriteAdds a comment to an alert or edits an existing one.
getUseCasesreadLists the data use cases defined in the account.
getUseCaseTableSummaryreadGets a summary of the tables that belong to a use case.
getUseCaseTablesreadLists the tables that belong to a use case.
createCustomDashboardWidgetwriteCreates a custom widget on a Monte Carlo dashboard.
testConnectionreadVerifies authentication and connectivity to Monte Carlo.
getDomainsreadLists the Monte Carlo domains in the account.
getUserreadGets information about the current user.
getCurrentTimereadReturns the current time, used to build time-range filters.

What Monte Carlo MCP exposes

Monte Carlo MCP groups its 24 tools into 5 categories, each with a suggested starting policy.

All 24 tools as JSON, ready to paste.

Category	Covers	Access	Suggested policy
Assets & lineage	Finding tables and assets by name, reading a table's schema, stats and metadata, its recent queries and the full SQL of a query, and tracing upstream and downstream lineage down to BI reports	Read	Read-only, but getQueryData returns full SQL text, which can carry business logic and literal values from the warehouse. Allow broadly for investigation agents and log which assets they read
Monitors	Reading monitor configurations and notification audiences, and generating monitors-as-code YAML for validation, metric, comparison and custom SQL monitors	Read	The Mac tools return YAML rather than changing live monitors, so applying it stays a separate, reviewable step through the Monte Carlo CLI. Safe to allow; the custom SQL variant is where to look first in review
Alerts	Reading active incidents and alerts, updating an alert, assigning its owner, and adding or editing alert comments	Read / Write	This is the main write surface. updateAlert changes the alert record that the people on the incident queue rely on, so gate it and setAlertOwner to triage agents and log every change. Comments are lower risk
Use cases & dashboards	Listing use cases and the tables behind them, summarizing a use case's tables, and creating custom dashboard widgets	Read / Write	Reads are safe. createCustomDashboardWidget adds content to dashboards other people use, so keep it to agents meant to build reporting
Account & utility	Testing the connection, listing domains, reading the current user and getting the current time	Read	Low-risk helpers that other calls depend on. Safe to allow everywhere

Connect Monte Carlo 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 Monte Carlo
Find Monte Carlo 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 Monte Carlo MCP or block it. You allow the exact tools each agent can call, one by one.

Monte Carlo MCP	24 TOOLS AVAILABLE
Code review agent	4 tools, read only
Issue triage bot	9 tools
Platform team assistant	all 24 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 Monte Carlo 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.	Monte Carlo MCP24 tools on the remote server at mcp.getmontecarlo.com/mcp over streamable HTTP, with OAuth 2.1 and dynamic client registration or MCP Server-type API keys. Tool list as published in the Claude connectors directory; Monte Carlo's docs now name tools in snake_case (get_alerts) and add extended and agent_observability toolsets through the x-mcd-toolset header.
Coding assistants	Connection & credentialsper user	Assets & lineage6 tools
Internal copilots	Tool access policiesdeny by default	Monitors6 tools
Workflow automations	Request & response inspectioninline	Alerts4 tools
Compatible MCP clients	Activity monitoringalerts and logs	All other tools8 tools