Загрузка видео...

Не удалось загрузить видео

На главную

# Three doors, one building: API, CLI, MCP API, CLI and MCP are three ways to connect an AI system to a tool. Picture one building with three entrances. 1. **API: the loading dock.** A program sends a precisely formatted request and receives structured data. Useful for bulk work,...

pro.glitch's profile picture

Professor Glitch

114,300 subscribers

22,500 просмотров • 1 месяц назад •via TikTok

ОбразованиеНаука и технологии#api#cli#mcp
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

CLI, API, MCP — the 3 ways your AI connects to everything (and when to use each) Your AI's brain is useless if it can't reach anything — the real power is in the connections. There are exactly 3 doors your AI can use, and once you know all three you can connect it to basically anything: DOOR 1 — CLI (the workshop) → typed commands. Commands are text, and text is what AI does best — so an AI with terminal access is a power user at your machine. And it's not just local files: GitHub, Stripe, Google Cloud, even your whole Google Workspace ship command-line tools. You say "pull last month's payments" — it types the commands. Job's on your computer or a tool has a CLI → this door. DOOR 2 — API (the service counter) → every serious online service has an official software entrance. Send a precise request, get an exact structured answer. Trade-off: each service needs its own wiring + key. Need ONE service with deep control → this door. DOOR 3 — MCP (the universal plug) → the standard built FOR AI. A service runs a small MCP server that announces its tools, and any AI can plug in — no custom wiring. Plug in once, the tools just show up. Want plug-and-play → this door. The real answer: a serious AI employee uses all three. They're not competitors — they're three doors on the same building, and you pick per job. Want to learn this whole world from scratch — AI, automations, agents, connecting everything, all the way to AI employees? That's exactly what I teach. Link in my profile. #aiagents #cli #mcp #api
3:34

CLI, API, MCP — the 3 ways your AI connects to everything (and when to use each) Your AI's brain is useless if it can't reach anything — the real power is in the connections. There are exactly 3 doors your AI can use, and once you know all three you can connect it to basically anything: DOOR 1 — CLI (the workshop) → typed commands. Commands are text, and text is what AI does best — so an AI with terminal access is a power user at your machine. And it's not just local files: GitHub, Stripe, Google Cloud, even your whole Google Workspace ship command-line tools. You say "pull last month's payments" — it types the commands. Job's on your computer or a tool has a CLI → this door. DOOR 2 — API (the service counter) → every serious online service has an official software entrance. Send a precise request, get an exact structured answer. Trade-off: each service needs its own wiring + key. Need ONE service with deep control → this door. DOOR 3 — MCP (the universal plug) → the standard built FOR AI. A service runs a small MCP server that announces its tools, and any AI can plug in — no custom wiring. Plug in once, the tools just show up. Want plug-and-play → this door. The real answer: a serious AI employee uses all three. They're not competitors — they're three doors on the same building, and you pick per job. Want to learn this whole world from scratch — AI, automations, agents, connecting everything, all the way to AI employees? That's exactly what I teach. Link in my profile. #aiagents #cli #mcp #api

Professor Glitch

12,200 просмотров • 3 месяцев назад

Ship your API to every MCP client in minutes 🚢 Upload your OpenAPI spec to Stainless → Generate an MCP Server → Your users can access your API in Claude Code Community, Cursor, and VS Code Watch me use the Replicate MCP Server to create a logo for my upcoming AI Podcast Generator, all with a single prompt! Now any MCP client can generate images and videos directly
2:57

Ship your API to every MCP client in minutes 🚢 Upload your OpenAPI spec to Stainless → Generate an MCP Server → Your users can access your API in Claude Code Community, Cursor, and VS Code Watch me use the Replicate MCP Server to create a logo for my upcoming AI Podcast Generator, all with a single prompt! Now any MCP client can generate images and videos directly

Adam Anzuoni

22,423 просмотров • 1 год назад

The rumors are true: Everything you’ve built in Salesforce is now an API, MCP tool, or CLI command with Headless 360.
0:34

The rumors are true: Everything you’ve built in Salesforce is now an API, MCP tool, or CLI command with Headless 360.

Salesforce

397,558 просмотров • 6 месяцев назад

Martin Keen from IBM just explained the debate that's splitting Claude and AI agent developers in half CLI vs MCP - and the answer will save you thousands of tokens > GitHub MCP server loads 80 tools into context = 55,000 tokens before your agent does anything > CLI: agent already knows grep, cat, git cold from training data > MCP wins when Claude needs to render a JavaScript page - curl can't do that, MCP browser server can in 250 tokens > MCP wins for Slack, Notion, databases - OAuth handled by the server, not the agent the rule: use CLI when commands map directly to the job, use MCP when the abstraction earns its cost full breakdown above
13:25

Martin Keen from IBM just explained the debate that's splitting Claude and AI agent developers in half CLI vs MCP - and the answer will save you thousands of tokens > GitHub MCP server loads 80 tools into context = 55,000 tokens before your agent does anything > CLI: agent already knows grep, cat, git cold from training data > MCP wins when Claude needs to render a JavaScript page - curl can't do that, MCP browser server can in 250 tokens > MCP wins for Slack, Notion, databases - OAuth handled by the server, not the agent the rule: use CLI when commands map directly to the job, use MCP when the abstraction earns its cost full breakdown above

Mr. Buzzoni

40,771 просмотров • 6 месяцев назад

New course: MCP: Build Rich-Context AI Apps with Anthropic. Learn to build AI apps that access tools, data, and prompts using the Model Context Protocol in this short course, created in partnership with Anthropic Anthropic and taught by Elie Schoppik Elie Schoppik, its Head of Technical Education. Connecting AI applications to external systems that bring rich context to LLM-based applications has often meant writing custom integrations for each use case. MCP is an open protocol that standardizes how LLMs access tools, data, and prompts from external sources, and simplifies how you provide context to your LLM-based applications. For example, you can provide context via third-party tools that let your LLM make API calls to search the web, access data from local docs, retrieve code from a GitHub repo, and so on. MCP, developed by Anthropic, is based on a client-server architecture that defines the communication details between an MCP client, hosted inside the AI application, and an MCP server that exposes tools, resources, and prompt templates. The server can be a subprocess launched by the client that runs locally or an independent process running remotely. In this hands-on course, you'll learn the core architecture behind MCP. You’ll create an MCP-compatible chatbot, build and deploy an MCP server, and connect the chatbot to your MCP server and other open-source servers. Here’s what you’ll do: - Understand why MCP makes AI development less fragmented and standardizes connections between AI applications and external data sources - Learn the core components of the client-server architecture of MCP and the underlying communication mechanism - Build a chatbot with custom tools for searching academic papers, and transform it into an MCP-compatible application - Build a local MCP server that exposes tools, resources, and prompt templates using FastMCP, and test it using MCP Inspector - Create an MCP client inside your chatbot to dynamically connect to your server - Connect your chatbot to reference servers built by Anthropic’s MCP team, such as filesystem, which implements filesystem operations, and fetch, which extracts contents from the web as markdown - Configure Claude Desktop to connect to your server and others, and explore how it abstracts away the low-level logic of MCP clients - Deploy your MCP server remotely and test it with the Inspector or other MCP-compatible applications - Learn about the roadmap for future MCP development, such as multi-agent architecture, MCP registry API, server discovery, authorization, and authentication MCP is an exciting and important technology that lets you build rich-context AI applications that connect to a growing ecosystem of MCP servers, with minimal integration work. Please sign up here!
2:43

New course: MCP: Build Rich-Context AI Apps with Anthropic. Learn to build AI apps that access tools, data, and prompts using the Model Context Protocol in this short course, created in partnership with Anthropic Anthropic and taught by Elie Schoppik Elie Schoppik, its Head of Technical Education. Connecting AI applications to external systems that bring rich context to LLM-based applications has often meant writing custom integrations for each use case. MCP is an open protocol that standardizes how LLMs access tools, data, and prompts from external sources, and simplifies how you provide context to your LLM-based applications. For example, you can provide context via third-party tools that let your LLM make API calls to search the web, access data from local docs, retrieve code from a GitHub repo, and so on. MCP, developed by Anthropic, is based on a client-server architecture that defines the communication details between an MCP client, hosted inside the AI application, and an MCP server that exposes tools, resources, and prompt templates. The server can be a subprocess launched by the client that runs locally or an independent process running remotely. In this hands-on course, you'll learn the core architecture behind MCP. You’ll create an MCP-compatible chatbot, build and deploy an MCP server, and connect the chatbot to your MCP server and other open-source servers. Here’s what you’ll do: - Understand why MCP makes AI development less fragmented and standardizes connections between AI applications and external data sources - Learn the core components of the client-server architecture of MCP and the underlying communication mechanism - Build a chatbot with custom tools for searching academic papers, and transform it into an MCP-compatible application - Build a local MCP server that exposes tools, resources, and prompt templates using FastMCP, and test it using MCP Inspector - Create an MCP client inside your chatbot to dynamically connect to your server - Connect your chatbot to reference servers built by Anthropic’s MCP team, such as filesystem, which implements filesystem operations, and fetch, which extracts contents from the web as markdown - Configure Claude Desktop to connect to your server and others, and explore how it abstracts away the low-level logic of MCP clients - Deploy your MCP server remotely and test it with the Inspector or other MCP-compatible applications - Learn about the roadmap for future MCP development, such as multi-agent architecture, MCP registry API, server discovery, authorization, and authentication MCP is an exciting and important technology that lets you build rich-context AI applications that connect to a growing ecosystem of MCP servers, with minimal integration work. Please sign up here!

Andrew Ng

142,494 просмотров • 1 год назад

I'm still picking my jaw off the floor with this MCP update: Postman will now automatically build an MCP server for an API. Automatically! You select your API and the requests you want to include, and Potman will generate a Zip file with everything you need. In collaboration with Postman, I recorded a video showing how I generated an MCP server for the Google Maps API. It's glorious! Also, and before I forget: You can now test any MCP server with your Postman application. They built MCP right in, so now you can: • Connect to an MCP server • List out all of the available tools • Test any of the tools This is a 10x boost for the MCP ecosystem. If you've tried the default MCP debugger, you'll understand how huge a deal it is that you can do all of that from Postman now. Go here to give this a try:
6:15

I'm still picking my jaw off the floor with this MCP update: Postman will now automatically build an MCP server for an API. Automatically! You select your API and the requests you want to include, and Potman will generate a Zip file with everything you need. In collaboration with Postman, I recorded a video showing how I generated an MCP server for the Google Maps API. It's glorious! Also, and before I forget: You can now test any MCP server with your Postman application. They built MCP right in, so now you can: • Connect to an MCP server • List out all of the available tools • Test any of the tools This is a 10x boost for the MCP ecosystem. If you've tried the default MCP debugger, you'll understand how huge a deal it is that you can do all of that from Postman now. Go here to give this a try:

Santiago

215,882 просмотров • 1 год назад

Introducing Spira MCP & CLI. Now you can connect Spira directly into the AI tools you already use. Create ready-to-post social videos from Claude, ChatGPT and more. Start with an idea, a script, or a product image. No switching between tools. No complicated production workflow. Get started with Local MCP > npx -y spira-lab/mcp CLI > npm install -g spira-lab/cli
0:43

Introducing Spira MCP & CLI. Now you can connect Spira directly into the AI tools you already use. Create ready-to-post social videos from Claude, ChatGPT and more. Start with an idea, a script, or a product image. No switching between tools. No complicated production workflow. Get started with Local MCP > npx -y spira-lab/mcp CLI > npm install -g spira-lab/cli

Spira AI

138,135 просмотров • 6 дней назад

Gemini CLI now integrates with FastMCP, Python's leading library for building MCP servers! Effortlessly connect your custom MCP tools and prompts, directly to Gemini CLI! Read the announcement→
0:23

Gemini CLI now integrates with FastMCP, Python's leading library for building MCP servers! Effortlessly connect your custom MCP tools and prompts, directly to Gemini CLI! Read the announcement→

Google Cloud Tech

29,494 просмотров • 1 год назад

MCP explained in one drawing: how your AI gets inside your apps (Claude, n8n, Gmail) Here's how your AI actually gets inside your apps. It's called MCP, and it fits in one drawing. On its own, an AI is a brain in a box. It can talk and think, but it can't open your email or build anything in n8n. It has no hands. 1. The door is the API. Most apps have a door for other software: Gmail has one, GitHub has one, n8n has one. Every door is a different shape, with its own rules and its own key. 2. The plug is MCP. It's one standard shape. On one side is the AI. On the other side is a small program called an MCP server, one per app, and it knows that app's door. The AI plugs in, and the server does the walking through the door. 3. The menu is the list of tools. When the AI plugs in, it sees every tool the app offers, in plain words. Gmail's server offers create a draft, search mail and read a message. n8n's server offers list workflows, create a workflow and test it. The AI reads the list like a menu, picks a tool and calls it. That's how Claude builds a whole workflow inside n8n, node by node, and checks its own work, while you never open n8n. 4. Connecting one takes one sentence. In Claude Code, say: "Connect yourself to my n8n, here's the address and a key I made just for you." Claude writes a small settings file with the server's name and the key. Type /mcp and it says connected. That's the setup for most apps. Some, like Gmail, you just sign in to instead. 5. The key is the important part. An MCP server can only use the tools on its menu, and only as far as its key allows. So make a key that can do the job and nothing more. If the app lets you limit a key, make it read only when reading is all you need. If it doesn't, use a test account. And if a key ever leaks, cancel it inside the app. One click, and nobody can use it, the AI included. One honest line: MCP doesn't make the AI smarter. It gives a smart thing a way to act, and everything it does through that plug, it does as you. So you still check the work, like you would a new hire's. If you're starting from scratch, I have step-by-step lessons for all of this inside my community. There are eight courses, in order, from the basics of AI to automations, building websites and apps, and AI agents. If you're interested, the link is in my profile. #mcp #cli ##aiagents #aiautomation #claudecode
2:56

MCP explained in one drawing: how your AI gets inside your apps (Claude, n8n, Gmail) Here's how your AI actually gets inside your apps. It's called MCP, and it fits in one drawing. On its own, an AI is a brain in a box. It can talk and think, but it can't open your email or build anything in n8n. It has no hands. 1. The door is the API. Most apps have a door for other software: Gmail has one, GitHub has one, n8n has one. Every door is a different shape, with its own rules and its own key. 2. The plug is MCP. It's one standard shape. On one side is the AI. On the other side is a small program called an MCP server, one per app, and it knows that app's door. The AI plugs in, and the server does the walking through the door. 3. The menu is the list of tools. When the AI plugs in, it sees every tool the app offers, in plain words. Gmail's server offers create a draft, search mail and read a message. n8n's server offers list workflows, create a workflow and test it. The AI reads the list like a menu, picks a tool and calls it. That's how Claude builds a whole workflow inside n8n, node by node, and checks its own work, while you never open n8n. 4. Connecting one takes one sentence. In Claude Code, say: "Connect yourself to my n8n, here's the address and a key I made just for you." Claude writes a small settings file with the server's name and the key. Type /mcp and it says connected. That's the setup for most apps. Some, like Gmail, you just sign in to instead. 5. The key is the important part. An MCP server can only use the tools on its menu, and only as far as its key allows. So make a key that can do the job and nothing more. If the app lets you limit a key, make it read only when reading is all you need. If it doesn't, use a test account. And if a key ever leaks, cancel it inside the app. One click, and nobody can use it, the AI included. One honest line: MCP doesn't make the AI smarter. It gives a smart thing a way to act, and everything it does through that plug, it does as you. So you still check the work, like you would a new hire's. If you're starting from scratch, I have step-by-step lessons for all of this inside my community. There are eight courses, in order, from the basics of AI to automations, building websites and apps, and AI agents. If you're interested, the link is in my profile. #mcp #cli ##aiagents #aiautomation #claudecode

Professor Glitch

18,500 просмотров • 26 дней назад

How to connect Agent Wallet to your AI agent. → MCP: add our MCP server to Claude, sign in with District Pass. → CLI: install the fdx CLI, sign up or log in via email OTP, run in Claude Code. Now your AI agent doesn't just talk. It pays, swaps, and earns. Get Started →
6:03

How to connect Agent Wallet to your AI agent. → MCP: add our MCP server to Claude, sign in with District Pass. → CLI: install the fdx CLI, sign up or log in via email OTP, run in Claude Code. Now your AI agent doesn't just talk. It pays, swaps, and earns. Get Started →

Finance District

18,293 просмотров • 5 месяцев назад

We’re excited to introduce the Unity CLI, our new terminal-native path for connecting coding agents, CI pipelines, and custom tooling to Unity. Unity MCP remains fully supported for your existing setups but to help with the transition, the CLI includes an "MCP Mode," allowing you to leverage existing MCP-based agents without starting from scratch. We want to remove barriers to innovation, so we are making these tools free! Both the CLI and MCP will be free, with no concurrency limits on the MCP server.
0:44

We’re excited to introduce the Unity CLI, our new terminal-native path for connecting coding agents, CI pipelines, and custom tooling to Unity. Unity MCP remains fully supported for your existing setups but to help with the transition, the CLI includes an "MCP Mode," allowing you to leverage existing MCP-based agents without starting from scratch. We want to remove barriers to innovation, so we are making these tools free! Both the CLI and MCP will be free, with no concurrency limits on the MCP server.

Unity

615,018 просмотров • 3 месяцев назад

Totally settling the "MCP vs CLI for coding agents" debate with this humongous blog post. I benchmarked screen, tmux, and a bespoke custom CLI and MCP tool against 3 task that require the agent to steer TUIs (lldb, python repl, opencode)
2:27

Totally settling the "MCP vs CLI for coding agents" debate with this humongous blog post. I benchmarked screen, tmux, and a bespoke custom CLI and MCP tool against 3 task that require the agent to steer TUIs (lldb, python repl, opencode)

Mario Zechner

65,131 просмотров • 1 год назад

this tweet misses the gold of executor! the whole point is you don't want CLI / scripts, you want API / MCP servers in this video i ask executor to configure itself and add an MCP server - it pops up a browser for me to OAuth through based on the MCP elicitation spec
1:50

this tweet misses the gold of executor! the whole point is you don't want CLI / scripts, you want API / MCP servers in this video i ask executor to configure itself and add an MCP server - it pops up a browser for me to OAuth through based on the MCP elicitation spec

Rhys

33,637 просмотров • 8 месяцев назад

What is the difference between CLI and MCP #cli #mcp #aiagent
2:29

What is the difference between CLI and MCP #cli #mcp #aiagent

Professor Glitch

10,000 просмотров • 5 месяцев назад

AI Agents can now connect to 100,000+ APIs instantly. Postman's AI Agent builder can turn any API into an MCP server in seconds without writing a single line of Code. You can use it in Claude, Cursor, Windsurf and other AI tools.
0:21

AI Agents can now connect to 100,000+ APIs instantly. Postman's AI Agent builder can turn any API into an MCP server in seconds without writing a single line of Code. You can use it in Claude, Cursor, Windsurf and other AI tools.

Shubham Saboo

79,921 просмотров • 1 год назад

Gemini CLI can automate your computer using MCP 🔥 Add Windows MCP (or macOS MCP) to Gemini CLI and you can tell it what to do autonomously. Gemini then takes control of your entire system to achieve the goal you've set. Links below
0:27

Gemini CLI can automate your computer using MCP 🔥 Add Windows MCP (or macOS MCP) to Gemini CLI and you can tell it what to do autonomously. Gemini then takes control of your entire system to achieve the goal you've set. Links below

Paul Couvert

132,010 просмотров • 1 год назад

Introducing Deep Graph Codebase to MCP with One Click... Now Available for Gemini CLI too! 🤩 Transform any public GitHub repository into an AI-ready Code Graph by simply changing the URL from to This instantly prepares the repository for seamless integration with AI assistants. It's that easy! You can now use repositories like ollama as a code graph within any project using Gemini CLI... just add the MCP Server JSON configuration to your .gemini/settings.json file
1:01

Introducing Deep Graph Codebase to MCP with One Click... Now Available for Gemini CLI too! 🤩 Transform any public GitHub repository into an AI-ready Code Graph by simply changing the URL from to This instantly prepares the repository for seamless integration with AI assistants. It's that easy! You can now use repositories like ollama as a code graph within any project using Gemini CLI... just add the MCP Server JSON configuration to your .gemini/settings.json file

Daniel San

53,516 просмотров • 1 год назад

The real difference between an API and an MCP One small detail makes the whole thing click: when an AI connects to a tool the new way, the tool talks first. The old way ran the internet for twenty years. Every serious app has a plug on the back — the API — and a programmer reads the manual, writes code that speaks its exact language, and wires two apps together by hand. Rock solid. But every move that connection can make was written ahead of time, by a person, and it repeats those moves forever. Then AI showed up — a worker that decides its next move LIVE, mid-job. It needed a way to walk up to a tool it has never seen and learn it on the spot. That's what MCP does. Connect an AI to your calendar and the calendar introduces itself first: here's what I can do — send an event, search the week, move a meeting — and here's what each one needs. The AI reads that card like a new employee reading the equipment list, picks the right tool, and fires. On its own. And the detail I love: under the hood, MCP still runs through the same plug. The API does the actual work. MCP adds the introduction on top, written so an AI can read it. That's how AI got hands inside real accounts — inbox, calendar, sheets. Log in, grant access, minutes. And tools that ship next year work the day they arrive, because they introduce themselves too. Keep one line: programs need instructions. Intelligence just needs an introduction. #api #mcp #aiautomation #aiagent
2:21

The real difference between an API and an MCP One small detail makes the whole thing click: when an AI connects to a tool the new way, the tool talks first. The old way ran the internet for twenty years. Every serious app has a plug on the back — the API — and a programmer reads the manual, writes code that speaks its exact language, and wires two apps together by hand. Rock solid. But every move that connection can make was written ahead of time, by a person, and it repeats those moves forever. Then AI showed up — a worker that decides its next move LIVE, mid-job. It needed a way to walk up to a tool it has never seen and learn it on the spot. That's what MCP does. Connect an AI to your calendar and the calendar introduces itself first: here's what I can do — send an event, search the week, move a meeting — and here's what each one needs. The AI reads that card like a new employee reading the equipment list, picks the right tool, and fires. On its own. And the detail I love: under the hood, MCP still runs through the same plug. The API does the actual work. MCP adds the introduction on top, written so an AI can read it. That's how AI got hands inside real accounts — inbox, calendar, sheets. Log in, grant access, minutes. And tools that ship next year work the day they arrive, because they introduce themselves too. Keep one line: programs need instructions. Intelligence just needs an introduction. #api #mcp #aiautomation #aiagent

Professor Glitch

14,100 просмотров • 2 месяцев назад

Who brought developers into this conversation? The disrespect is wild. An average designer with the right Codex, Claude, or Cursor setup, .md file, plus a CLI and MCP server, can literally make some of you irrelevant. Let that sink in.
0:25

Who brought developers into this conversation? The disrespect is wild. An average designer with the right Codex, Claude, or Cursor setup, .md file, plus a CLI and MCP server, can literally make some of you irrelevant. Let that sink in.

TimX | Million-Dollar UX Design 🚀

111,189 просмотров • 7 месяцев назад