Loading video...

Video Failed to Load

Go Home

Introducing the One Remote MCP. 592+ apps. 94,357+ tools. 4 tools in your context and a flat 3,000 tokens. It's the MCP that connects your AI to everything. One URL, nothing to install. Try it out now:

24,170 views • 1 month ago •via X (Twitter)

29 Comments

Moe's profile picture
Moe1 month ago

Here's what a normal setup costs before you type a word: - jira: 17,000 tokens - playwright: 14,000 - slack: 14,000 - shopify: 10,000 - airtable: 8,000 5 servers, 55,000 tokens of tool definitions your agent hasn't called yet.

Moe's profile picture
Moe1 month ago

One exposes 4 operations, no matter how many apps you connect. > list integrations > search actions > get knowledge > execute action 3,000 tokens with 1 app connected or all 592.

Moe's profile picture
Moe1 month ago

An action's docs load the second your agent calls it. So 94,357 tools stay one search away instead of sitting in the window doing nothing, and your 50th connection doesn't move the number.

Moe's profile picture
Moe1 month ago

There's nothing to install and no API key to manage. Paste " into Claude, ChatGPT, Cursor, or VS Code, sign in with OAuth, and every app you've connected is there.

Moe's profile picture
Moe1 month ago

Then decide what your agent gets. The consent screen scopes it: which project, which environment, and per connection, read only, full, or a specific list of actions. We enforce that on our side on every call.

Moe's profile picture
Moe1 month ago

Knowledge mode removes execution completely. The agent still reads every schema, auth flow, and caveat we tested on the real API. It just can't fire a live request. That's the mode for Cursor, where it's writing your integration code.

Moe's profile picture
Moe1 month ago

One URL between your agent and everything your users already pay for. Try it out now:

You.com's profile picture
You.com1 month ago

👏

GARLOTIC's profile picture
GARLOTIC1 month ago

this is what mcp should feel like. less context bloat, fewer tokens wasted, and everything available when you actually need it. very clean.

Moe's profile picture
Moe1 month ago

🫡🫡

Jaisal Kothari's profile picture
Jaisal Kothari1 month ago

This is huge, congrats guys!!

Moe's profile picture
Moe1 month ago

Thank you sir

Chesny's profile picture
Chesny1 month ago

The flat 3,000 tokens is the part that matters Most setups burn 50k on tool definitions the agent never calls

Moe's profile picture
Moe1 month ago

You are spot on, spot on!!

Nico's profile picture
Nico1 month ago

This is huuuuge

Moe's profile picture
Moe1 month ago

🔥🙌can't wait to hear your feedback

Muhammad Ayan's profile picture
Muhammad Ayan1 month ago

Knowledge mode for Cursor is lowkey the smartest part of this setup :)

Moe's profile picture
Moe1 month ago

Totally agree! Looking forward to hearing your feedback

astro's profile picture
astro1 month ago

the approach of exposing only 4 tools and loading the documentation on-demand is exactly what was needed, the context bloat of traditional MCPs was unsustainable.

xIA's profile picture
xIA1 month ago

really amazing guys

Moe's profile picture
Moe1 month ago

🔥

Wei佳's profile picture
Wei佳1 month ago

94k tools behind four context tools is the useful abstraction; the hard part is keeping discovery and permissions predictable as the catalog grows.

Moe's profile picture
Moe1 month ago

You nailed it.

Dipti Sharma's profile picture
Dipti Sharma1 month ago

The 4-tool context limit is a smart design choice.

Moe's profile picture
Moe1 month ago

We believe so too! Thank you 🙏

Rony's profile picture
Rony1 month ago

Huge congrats on the launch! Excited to see how developers build with this.

Moe's profile picture
Moe1 month ago

Looking forward to hearing your feedback!

Harsh Makadia's profile picture
Harsh Makadia1 month ago

This is a solid feature, MCPs are the future! 🔥 Congratulations on the launch Moe!

Moe's profile picture
Moe1 month ago

Much appreciated!!

Related Videos

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

142,234 views • 1 year ago