# Zoom APIs vs. MCP

<!-- WARNING: THIS FILE WAS AUTOGENERATED! DO NOT EDIT! -->

## When to use which

Zoom APIs and model context protocol (MCP) solve different, but complementary, problems. Choosing the right one upfront helps you move faster and avoid unneeded complexity.

## Use Zoom APIs when you need direct platform control

Zoom APIs are best when you're building deterministic, system-driven integrations. Use APIs if you want to:

-   Retrieve or update Zoom data programmatically (meetings, users, phone settings, calendars, mail, workspaces).
-   Automate account or user configuration at scale.
-   Build backend services, admin tooling, or scheduled jobs.
-   Maintain precise control over request logic, error handling, and retries.

### Typical examples

-   Provisioning users and licenses automatically.
-   Pulling call logs or meeting metrics into a data warehouse.
-   Updating account settings or phone assignments programmatically.
-   Building internal tools that require predictable, repeatable behavior.

### Why this matters

APIs give you full control and clarity. They are ideal when correctness, repeatability, and explicit logic matter more than flexibility.

## Use MCP when you're building AI-driven or tool-based experiences

MCP is designed for AI systems that need to discover and use tools dynamically. Use MCP if you want to:

-   Expose tools or data to custom agents or Zoom AI Companion within Zoom AI Studio.
-   Let AI agents choose when and how to call tools based on context.
-   Access Zoom data from external AI platforms using a standard interface.
-   Avoid building custom connectors for each AI system.

### Typical examples

-   Allowing AI Companion to query internal knowledge bases or third-party systems.
-   Enabling an external AI platform to retrieve Zoom meeting summaries or chat context.
-   Building AI workflows that span Zoom and other MCP-compliant services.
-   Rapidly iterating on AI capabilities without rewriting integrations.

### Why this matters

MCP optimizes for flexibility and interoperability. It lets AI reason about available tools instead of hard-coding integrations.

## Use APIs and MCP together for the best results

In many real-world systems, APIs and MCP work best in combination.

### A common pattern

-   Use **Zoom APIs** to handle core platform access, configuration, and automation.
-   Use **MCP** to expose higher-level tools and context to AI systems.

### For example

-   APIs manage user provisioning and data ingestion.
-   MCP exposes curated tools that AI Companion or external agents can invoke safely.

### Why this matters

This layered approach lets you keep your core systems stable while enabling rapid AI utility on top.

## Quick decision guide

If your primary goal is:

-   **Automation, configuration, or reporting** → use **Zoom APIs**
-   **AI interaction, tool discovery, or cross-platform AI workflows** → use **MCP**
-   **Enterprise AI experiences at scale** → use **both**

In essence, Zoom APIs provide **control, reliability, and scale** while MCP provides **intelligence, flexibility, and interoperability**. This lets teams move from traditional integrations to AI-powered experiences without re-architecting their foundation.