User Guide

ApexMCP is an MCP (Model Context Protocol) gateway. It connects your AI agents to your data sources — databases, APIs, file storage — without you writing any glue code.

What You Can Do

Task	Where
Connect a database, API, or file bucket	Connectors
Give your AI agent access to those tools	MCP Endpoint & API Keys
Test tools in the browser	MCP Endpoint → Tool Tester
Invite teammates and assign roles	Team Management
View usage and upgrade your plan	Billing & Plans
Rotate API keys, manage org settings	Settings

How It Works

Add a connector — you tell ApexMCP how to reach your data source (host, credentials, auth type). Credentials are stored encrypted in a vault.
ApexMCP generates MCP tools — e.g. connecting a PostgreSQL database automatically generates a query_postgres_<name> tool for each table/view the user has SELECT on.
Your agent calls tools — point any MCP-compatible AI agent at your endpoint URL with an API key. It calls tools/list to discover available tools and tools/call to execute them.

Supported Data Sources

Type	What it does
PostgreSQL	Query tables and views
MySQL	Query tables and views
MongoDB	Find documents, run aggregation pipelines
REST API	Call endpoints (Basic, OAuth2, JWT auth)
GraphQL API	Run queries against any GraphQL schema
AWS S3 / S3-compatible	List and retrieve objects
Google Cloud Storage	List and retrieve objects
Google Sheets	Read sheet data
Inbound Webhook	Receive and replay events from external systems
External MCP Server	Proxy calls to your own MCP server

Roles

Access to features depends on your role within the organisation.

Role	Access
Admin	Full access — connectors, team, billing, settings, audit logs
Finance	Billing, logs, dashboard
Contributor	Connectors, MCP endpoint, test tool
Pending	Onboarding only, no dashboard access until approved

Admins can invite members and assign roles from Team Management.

Introduction Dashboard Overview