How is this different from connecting Tinybird 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 Tinybird MCP API changes?

We track the Tinybird 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 Tinybird 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 Tinybird?

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 Tinybird MCP server myself?

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

Connect Tinybird MCP to your AI agents, securely.

Q: Is Tinybird MCP secure?

The server itself exposes 6 tools, and some of them can write to your Tinybird 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 Tinybird.

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

Get started for free

Get a demo

Try Tinybird MCP now

What your agents can ask once Tinybird MCP is connected.

Select the tools your AI agents can call

6 tools from the Tinybird MCP schema.

Showing 6 tools

list_datasourcesreadLists the Data Sources in the Workspace.
list_service_datasourcesreadLists the Organization and Tinybird service Data Sources available to the Workspace, which hold operational data such as query logs and consumption.
list_endpointsreadLists the published Endpoints with their names, parameters and descriptions, which is how an agent finds the endpoint name and arguments for a call.
call_endpointreadCalls a published Endpoint by name with an optional params object and a response format of CSVWithNames, JSON or Markdown. It respects the Endpoint's rate limits and the Token's scopes, and rejects parameters the Endpoint does not define with an error listing the accepted ones.
text_to_sqlreadConverts a natural-language question into a SQL query. It drafts the SQL rather than running it.
execute_queryhigh riskRuns arbitrary SQL against the Tinybird Query API, with an optional response format and an explain flag that returns the query plan instead of executing. What it can reach is defined by the Token's scopes rather than by the tool.

What Tinybird MCP exposes

Tinybird MCP groups its 6 tools into 3 categories, each with a suggested starting policy.

All 6 tools as JSON, ready to paste.

Category	Covers	Access	Suggested policy
Discovery	Listing the Data Sources in the Workspace, the Organization and Tinybird service Data Sources available to it, and the published Endpoints with their names, parameters and descriptions	Read	Read-only and the prerequisite for a valid endpoint call. list_service_datasources reaches Tinybird's own operational tables, which hold query logs and consumption data rather than the organization's business data
Endpoints	Calling a published Endpoint by name with its own parameters and an optional response format of CSV with names, JSON or Markdown, respecting the Endpoint's rate limits and the Token's scopes	Read	This is the safer half of the server, because an Endpoint is a query somebody already reviewed and published. Parameters the Endpoint does not define are rejected with a list of the accepted ones, and a JWT can pin which Endpoints are reachable at all
Query	Converting a natural-language question into SQL, and running SQL against the Tinybird Query API with an optional response format, an explain flag that returns the query plan instead of executing, and EXPLAIN options such as indexes and actions	Read / Write	execute_query runs arbitrary SQL, so its reach is set by the Token rather than by the tool, and the SQL text is what belongs in the log. The explain flag is a genuine dry run worth using before an expensive query, and text_to_sql only drafts SQL without running it

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

Tinybird MCP	6 TOOLS AVAILABLE
Code review agent	4 tools, read only
Issue triage bot	9 tools
Platform team assistant	all 6 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 Tinybird 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.	Tinybird MCP6 tools on the remote server at mcp.tinybird.co over streamable HTTP. Which tools are exposed depends on the Token: an admin Token exposes all of them, a resource-scoped static Token only the resources it permits, and a JWT can restrict access to specific Endpoints, add fixed parameters, filter Data Sources and set rate limits. The token is passed as a query parameter in the connection URL. Every MCP request carries from=mcp and is visible in the service Data Sources.
Coding assistants	Connection & credentialsper user	Discovery3 tools
Internal copilots	Tool access policiesdeny by default	Query2 tools
Workflow automations	Request & response inspectioninline	Endpoints1 tool
Compatible MCP clients	Activity monitoringalerts and logs