MCP server Databases terbaik untuk Codex

Connects AI agents and IDEs to enterprise databases using prebuilt tools or a custom framework for secure, structured queries.

[mcp_servers.mcp-toolbox]
command = "docker"
args = ["run", "-i", "--rm", "us-central1-docker.pkg.dev/database-toolbox/toolbox/toolbox:1.17.0"]

Connects AI agents and IDEs to enterprise databases using prebuilt tools or a custom framework for secure, structured queries.

[mcp_servers.genai-toolbox]
command = "docker"
args = ["run", "-i", "--rm", "us-central1-docker.pkg.dev/database-toolbox/toolbox/toolbox:0.31.0"]

Allows AI agents to query over 100 different databases using pre-configured connections.

[mcp_servers.dbx]
command = "npx"
args = ["-y", "@dbx-app/mcp-server"]

Maps lab results to LOINC codes and UCUM units and symptoms to ICPC-3 codes for AI agents, offline and without an API key.

[mcp_servers.mirobody]
command = "uvx"
args = ["mirobody"]

Exposes your Nhost GraphQL API as an MCP server to allow AI assistants to query and mutate project data within user-defined permissions.

[mcp_servers.mcp]
command = "docker"
args = ["run", "-i", "--rm", "docker.io/nhost/mcp:0.4.0-beta2"]

Keeps bulky tool output out of a coding agent's context window by running work in a local sandbox and indexing results in SQLite for on-demand search.

[mcp_servers.context-mode]
command = "npx"
args = ["-y", "context-mode"]

Provides an open-source China A-share data lake that allows AI agents to query stock prices, fundamentals, flows, and derivatives via MCP.

[mcp_servers.cnequity]
command = "uvx"
args = ["cnequity"]

Lets AI clients search, explore, query, and visualize data within a Metabase instance using the Model Context Protocol.

[mcp_servers.mcp]
url = "https://{metabase_host}/api/mcp"

Allows AI agents to interact with Couchbase clusters and Operational Insights to manage data, query schemas, and monitor cluster health.

[mcp_servers.mcp-server-couchbase]
command = "uvx"
args = ["couchbase-mcp-server"]

Query and explore ClickHouse clusters and chDB embedded databases with SQL from an AI client.

[mcp_servers.mcp-clickhouse]
command = "uvx"
args = ["mcp-clickhouse"]

Gives AI clients real-time Valkey and Redis monitoring, slow-command analysis, anomaly detection, and trace investigation.

[mcp_servers.monitor-9c1b20]
command = "npx"
args = ["-y", "@betterdb/mcp"]

Converts plain-English questions into SQL queries against your database schemas and returns the generated SQL and results.

[mcp_servers.queryweaver]
command = "docker"
args = ["run", "-i", "--rm", "docker.io/falkordb/queryweaver:0.0.11"]

Connect AI assistants to DuckDB, MotherDuck, or S3-hosted databases to execute SQL queries and perform data engineering tasks.

[mcp_servers.mcp-server-motherduck]
command = "uvx"
args = ["mcp-server-motherduck"]

Lets AI assistants run Apex, query records with SOQL, inspect metadata, and manage org connections through the Salesforce CLI.

[mcp_servers.salesforce-mcp-server]
command = "npx"
args = ["-y", "@advanced-communities/salesforce-mcp-server"]

Enables AI development tools to manage Firebase projects, query databases, manage authentication, and debug Crashlytics issues.

[mcp_servers.firebase-mcp]
command = "npx"
args = ["-y", "firebase-tools"]

Lets an AI agent create, modify, document, and validate Power BI semantic models, including tables, measures, and DAX queries.

[mcp_servers.powerbi-modeling-mcp]
command = "npx"
args = ["-y", "@microsoft/powerbi-modeling-mcp"]

Connects Firebase projects to FireCMS, infers collection schemas from Firestore data, and manages CMS data, users and configuration.

[mcp_servers.mcp]
command = "npx"
args = ["-y", "@firecms/mcp-server"]

Enables AI agents to discover datasets, inspect schemas, read spatial features, and execute sandboxed SQL queries against a GeoLens instance.

[mcp_servers.geolens]
command = "uvx"
args = ["geolens-mcp"]

Gives AI clients 68 tools for live crypto derivatives data: funding rates, arbitrage, open interest, liquidations, options and ETF flows.

[mcp_servers.bykaranteli-mcp]
command = "npx"
args = ["-y", "bykaranteli-mcp"]