Sign in Book a demo
Back to Tools
Databases API Key
Build Your Agent View YugabyteDB Docs
yugabytedb logo

YugabyteDB MCP Tools for AI Agents

Let an agent inspect YugabyteDB schemas, count rows, and run read-only SQL queries, returning database results as JSON.

Available Tools

2 tools available in this integration

Yugabytedb.RunReadOnlyQuery

Run a read-only SQL query against YugabyteDB and return results as JSON.

Yugabytedb.SummarizeDatabase

List all tables in a YugabyteDB database, including column schemas and row counts.

Why agent-optimized tools matter

Built for intent, not API schemas

Ask the agent to "find the tables that contain customer orders and summarize last month's totals." Arcade picks the right YugabyteDB tools without making you provide table names or schemas first.

Up to 100x fewer tokens

YugabyteDB schema summaries return compact table metadata, while query results contain only the columns and rows requested, so the agent need not load an entire database into context.

Complete the workflow end to end

Inspect the schema, identify relevant tables, run a read-only query, check the returned rows, and summarize the result through one toolkit designed around the complete database investigation task.

Make actions predictable and controlled

Arcade tools return structured results and useful errors across users and runs. Teams can apply scope restrictions, redaction, and approval policies so agents follow the same controls every time.

Securing YugabyteDB

The more capabilities agents get, the more useful they are. However, more capabilities bring more risks of undesired effects such as data loss and exfiltration.

This is why Arcade provides you with contextual rules that dictate how and when a tool is allowed to run. Arcade’s contextual access lets you govern tool visibility and behavior at every stage of execution and control who can see a tool, what inputs are allowed, and what comes back.

Learn about Contextual Access Policies

For YugabyteDB, you can implement some of these policies:

Access when tools are listed for a user
  • Decide which tools an agent is allowed to see at all. A tool that is never listed cannot be called, whoever is asking.
Pre-execution before each tool call
  • Inspect the inputs a call is about to run with, and block it or send it for human approval when they fall outside what the agent should do unsupervised.
Post-execution after the tool returns
  • Redact Yugabytedb.RunReadOnlyQuery results containing passwords, API keys, access tokens, or direct customer identifiers before they reach the model, and log each redaction.
  • Strip table and column names that reveal credentials, payment data, or customer records from Yugabytedb.SummarizeDatabase results before the model sees the schema.

How to use YugabyteDB with your agent

Connect YugabyteDB to an MCP client and follow the setup from authentication through your first database query.

Your questions, answered

A YugabyteDB MCP server exposes database capabilities as structured tools that an AI agent can call through the Model Context Protocol. With Arcade, an agent can inspect database structure and run read-only SQL queries that return JSON results.

YugabyteDB does not provide a first-party MCP server in this integration. Arcade provides the MCP gateway, shared-key authentication, tool execution, approval and redaction policies, and access from supported clients.

Arcade agents can list all tables with their column schemas and row counts, then run read-only SQL queries against YugabyteDB and receive the results as JSON. They cannot insert, update, delete, or otherwise modify database data through these actions.

Arcade uses one shared API key or secret configured for the integration. That credential must have access to the YugabyteDB database and permissions sufficient for metadata inspection and read-only SQL; write permissions are not needed for these actions.

No. Users do not connect separate YugabyteDB accounts through this integration. One shared API key or secret is configured once, brokered by Arcade, and used for calls made through that connection.

Yes. Contextual Access policies can require human approval before a database call proceeds, and can redact returned fields or hide the actions from an agent. The available YugabyteDB actions are read-only, so they do not perform consequential database writes.

Arcade proxies YugabyteDB API calls and does not persist your database content as part of normal tool execution. The shared credential is brokered and stored encrypted at rest, and the model never receives it.

Any framework or client that can connect to Arcade's MCP gateway or tool APIs works, including LangChain, OpenAI Agents, CrewAI, Claude Desktop, Claude Code, Cursor, Visual Studio Code, Microsoft Copilot Studio, and GitHub Copilot.

Arcade uses the configured YugabyteDB credential while it remains valid. If the key or secret is revoked, rotated, or can no longer be used, calls stop until the credential is reconfigured.

Calls are subject to the limits and capacity of your YugabyteDB deployment and the configured API access path. These actions do not add write operations or a documented batch limit; query size and execution time still depend on the database configuration.

Ready to actually ship your agent?

Connect your agent to YugabyteDB with a shared API key, agent-optimized tools and approval controls for consequential actions.

Build Your Agent Read the YugabyteDB Quickstart