Genai Toolbox
VerifiedOpen source MCP server specializing in easy, fast, and secure tools for Databases.
Install
No auto-install command available for this server.
Check the GitHub repository for setup instructionsSafety Report
details
Integrated authentication layer for database connections; supports IAM, service accounts, and credential-based auth for Google Cloud services
details
HTTP-based MCP transport available; CORS settings configurable in deployment
details
Connection pooling provided; no explicit MCP-layer rate limiting documented — relies on database-level limits
details
No known CVEs; official Google open source project with security review process
details
Maintained by Google/googleapis org; dependencies subject to Google's internal security review processes
details
No dangerous code patterns detected
details
Apache 2.0 License
details
Open source on GitHub (googleapis/genai-toolbox); fully auditable
details
Actively maintained by Google; currently in beta with roadmap to v1.0. Regular releases.
Compatibility
Tools Provided (12)
execute_sql (postgres-sql) Executes a SQL query against a configured PostgreSQL database and returns results. Tool name is defined in tools.yaml configuration.
execute_sql (alloydb) Executes SQL queries against Google Cloud AlloyDB (PostgreSQL-compatible). Configured via tools.yaml with connection pooling and IAM auth.
execute_sql (cloud-sql-postgres/mysql/sqlserver) Executes SQL queries against Google Cloud SQL instances (PostgreSQL, MySQL, or SQL Server variants).
execute_sql (bigquery) Runs BigQuery SQL queries including analytics, ML functions, and large-scale data aggregations. Supports conversational analytics and forecasting.
execute_sql (spanner) Executes SQL or DML statements against Google Cloud Spanner distributed database instances.
execute_query (firestore) Queries Google Cloud Firestore documents using collection filters, with support for insert, update, and delete operations.
execute_sql (mysql/postgresql/sqlite) Executes SQL queries against self-hosted MySQL, PostgreSQL, or SQLite databases. Tool name and parameters defined in tools.yaml.
execute_query (mongodb) Runs MongoDB queries and aggregation pipelines against configured MongoDB collections.
execute_cypher (neo4j) Executes Cypher graph query language statements against a Neo4j graph database.
execute_command (redis/valkey) Runs Redis or Valkey commands (GET, SET, HGETALL, etc.) against configured in-memory data stores.
execute_sql (snowflake/bigquery/clickhouse) Executes analytical SQL queries against Snowflake, ClickHouse, or other OLAP-oriented database backends.
list_tables Lists available tables, schemas, or collections in the configured database. Enables schema discovery for query planning.
Frequently Asked Questions
What is Genai Toolbox?
Open source MCP server specializing in easy, fast, and secure tools for Databases.
What tools does Genai Toolbox provide?
Genai Toolbox provides 12 tools including execute_sql (postgres-sql), execute_sql (alloydb), execute_sql (cloud-sql-postgres/mysql/sqlserver), execute_sql (bigquery). Each tool can be called by AI agents to perform specific operations.
Is Genai Toolbox safe to use?
Yes, Genai Toolbox has passed our 8-point security audit and is rated Verified. All checks including authentication, dependency audit, and known CVE scan passed.
What are alternatives to Genai Toolbox?
Similar MCP servers include Kubeshark MCP Server, Supabase MCP, Postgres MCP. Each serves a similar purpose but may differ in features, language, and compatibility.
Similar MCP Servers
Kubeshark MCP Server
VerifiedReal-time Kubernetes network traffic visibility and API analysis for HTTP, gRPC, Redis, Kafka, DNS.
Supabase MCP
VerifiedOfficial Supabase MCP server to connect AI assistants directly with your Supabase project and allows them to perform tasks like managing tables, fetching config, and querying data.
Postgres MCP
VerifiedAll-in-one MCP server for Postgres development and operations, with tools for performance analysis, tuning, and health checks