Agent Academy is a free, source-linked library of skills, prompts, MCP servers and bot templates, organised into courses by profession, for people who hand real work to AI agents: 1,467 live resources, each with its original link, license and safety tags.
Skill
choose-zoom-approach
Original text · EnglishChoose the right Zoom architecture for a use case. Use when deciding between REST API, Webhooks, WebSockets, Meeting SDK, Video SDK, Zoom Apps SDK, Zoom MCP, Phone, Contact Center, or a hybrid approach.
Republished in full under its license, with attribution. The source is authoritative.
name: choose-zoom-approach
description: Choose the right Zoom architecture for a use case. Use when deciding between REST API, Webhooks, WebSockets, Meeting SDK, Video SDK, Zoom Apps SDK, Zoom MCP, Phone, Contact Center, or a hybrid approach.
user-invocable: false
Choose Zoom Approach
Pick the smallest correct Zoom surface for the job, then layer in only the supporting pieces that are actually required.
Decision Framework
Problem Type
Primary Zoom Surface
Deterministic backend automation, account management, reporting, scheduled jobs
[rest-api](../rest-api/SKILL.md)
Event delivery to your backend
[webhooks](../webhooks/SKILL.md) or [websockets](../websockets/SKILL.md)
Embed Zoom meetings into your app
[meeting-sdk](../meeting-sdk/SKILL.md)
Build a fully custom video experience
[video-sdk](../video-sdk/SKILL.md)
Build inside the Zoom client
[zoom-apps-sdk](../zoom-apps-sdk/SKILL.md)
AI-agent tool workflows over Zoom data
[zoom-mcp](../zoom-mcp/SKILL.md)
Real-time media extraction or meeting bots
[rtms](../rtms/SKILL.md) plus [meeting-sdk](../meeting-sdk/SKILL.md) when needed
Phone workflows
[phone](../phone/SKILL.md)
Contact Center or Virtual Agent flows
[contact-center](../contact-center/SKILL.md) or [virtual-agent](../virtual-agent/SKILL.md)
Guardrails
Do not recommend Video SDK when the user actually needs Zoom meeting semantics.
Do not recommend Meeting SDK when the user needs a fully custom session product.
Do not replace deterministic backend automation with MCP-only guidance.
Prefer hybrid rest-api + zoom-mcp when the user needs both stable system actions and AI-driven discovery.
Reference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`,…
Guide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar…
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
Use after generating code, after accepting AI suggestions, or when reviewing AI-written modules. Also use when code works but feels brittle, when error handling seems thin, when orphaned resources or missing cleanup are suspected, or when the agent claims done but hidden debt may exist. Catches the specific failure patterns AI agents produce that humans would not.