惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

月光博客
月光博客
Cyberwarzone
Cyberwarzone
L
LINUX DO - 最新话题
N
News and Events Feed by Topic
T
Troy Hunt's Blog
Help Net Security
Help Net Security
S
Security @ Cisco Blogs
Google DeepMind News
Google DeepMind News
Security Archives - TechRepublic
Security Archives - TechRepublic
M
MIT News - Artificial intelligence
G
Google Developers Blog
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
V2EX - 技术
V2EX - 技术
Y
Y Combinator Blog
D
Darknet – Hacking Tools, Hacker News & Cyber Security
大猫的无限游戏
大猫的无限游戏
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
Microsoft Security Blog
Microsoft Security Blog
Cisco Talos Blog
Cisco Talos Blog
T
Threatpost
Recent Commits to openclaw:main
Recent Commits to openclaw:main
S
SegmentFault 最新的问题
I
InfoQ
H
Hacker News: Front Page
D
Docker
Scott Helme
Scott Helme
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
Blog — PlanetScale
Blog — PlanetScale
人人都是产品经理
人人都是产品经理
博客园 - 叶小钗
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
N
Netflix TechBlog - Medium
AWS News Blog
AWS News Blog
Know Your Adversary
Know Your Adversary
博客园 - 【当耐特】
T
Tor Project blog
U
Unit 42
H
Heimdal Security Blog
Microsoft Azure Blog
Microsoft Azure Blog
K
KPMG report finds enterprise disconnect between AI and its ROI | CIO
P
Privacy & Cybersecurity Law Blog
PCI Perspectives
PCI Perspectives
美团技术团队
O
OpenAI News
T
Tailwind CSS Blog
H
Hackread – Cybersecurity News, Data Breaches, AI and More
B
Blog
GbyAI
GbyAI
cs.CL updates on arXiv.org
cs.CL updates on arXiv.org
MyScale Blog
MyScale Blog

Vercel News

Vercel Open Source Program: Winter 2026 cohort How Notion Workers run untrusted code at scale with Vercel Sandbox How we run Vercel's CDN in front of Discourse From idea to secure checkout in minutes with Stripe Building Slack agents can be easy Scaling redirects to infinity on Vercel Advancing Python typing Gamma builds design-first agents with Vercel How Avalara turns pipe dreams into patent-pending with v0 Keeping community human while scaling with agents How OpenEvidence built a healthcare AI that physicians actually trust Security boundaries in agentic architectures Skills Night: 69,000+ ways agents are getting smarter Video Generation with AI Gateway We Ralph Wiggumed WebStreams to make them 10x faster How Stably ships AI testing agents in hours, not weeks How we built AEO tracking for coding agents Anyone can build agents, but it takes a platform to run them Introducing Geist Pixel The Vercel AI Accelerator is back with $6m in credits Making agent-friendly pages with content negotiation The Vercel OSS Bug Bounty program is now available Introducing the new v0 Run untrusted code with Vercel Sandbox, now generally available How Stripe built a game-changing app in a single flight with v0 How Sensay went from zero to product in six weeks AGENTS.md outperforms skills in our agent evals Agent skills explained: An FAQ Testing if "bash is all you need" AWS databases are now live on the Vercel Marketplace and v0 Use Perplexity Web Search with Vercel AI Gateway Introducing: React Best Practices Nick Bogaty joins Vercel as Chief Revenue Officer How Mux shipped durable video workflows with their @mux/ai SDK How to build agents with filesystems and bash How we made v0 an effective coding agent Stopping the slow death of internal tools Building AI-Generated Pixel Trading Cards with Vercel AI Gateway We removed 80% of our agent’s tools AI SDK 6 Our $1 million hacker challenge for React2Shell Cline now runs on Vercel AI Gateway How to prompt v0 Build smarter workflows with Notion and v0 Vercel launches partner certification Inside Workflow DevKit: How framework integrations work React2Shell Security Bulletin | Vercel Knowledge Base Billions of requests: Black Friday-Cyber Monday 2025 Investing in the Python ecosystem AWS Databases coming to the Vercel Marketplace How we built the v0 iOS app Workflow Builder: Build your own workflow automation platform Vercel Open Source Program: Fall 2025 cohort Self-driving infrastructure Vercel collaborates with Google for Gemini 3 Pro Preview launch Vercel: The anti-vendor-lock-in cloud How Nous Research used BotID to block automated abuse at scale How AI Gateway runs on Fluid compute What we learned building agents at Vercel Build and deploy data applications on Snowflake with v0 BotID Deep Analysis catches a sophisticated bot network in real-time Vercel achieves TISAX AL2 compliance to serve automotive partners Bun runtime on Vercel Functions David Totten Joins Vercel to Lead Global Field Engineering Vercel Ship AI 2025 recap You can just ship agents AI agents and services on the Vercel Marketplace Built-in durability: Introducing Workflow Development Kit Zero-config backends on Vercel AI Cloud Introducing Vercel Agent: Your new Vercel teammate Update regarding Vercel service disruption on October 20, 2025 Agents at work, a partnership with Salesforce and Slack Running Next.js in ChatGPT: How to Build ChatGPT Apps Talha Tariq joins Vercel as CTO of Security Just another (Black) Friday Server rendering benchmarks: Fluid Compute and Cloudflare Workers Towards the AI Cloud: Our Series F Collaborating with Anthropic on Claude Sonnet 4.5 to power intelligent coding agents Preventing the stampede: Request collapsing in the Vercel CDN BotID uncovers hidden SEO poisoning How we made global routing faster with Bloom filters What you need to know about vibe coding Scale to one: How Fluid solves cold starts Addressing security & quality issues with MCP tools - Vercel AI agents at scale: Rox’s Vercel-powered revenue operating system Agentic Infrastructure Zero Data Retention on AI Gateway Optimizing Vercel Sandbox snapshots How Waldium made a blog platform work for humans and AI alike How FLORA shipped a creative agent on Vercel's AI stack Agent responsibly Making Turborepo 96% faster with agents, sandboxes, and humans Unified reporting for all AI Gateway usage new.website joins forces with v0 SERHANT.'s playbook for rapid AI iteration Two startups at global scale without DevOps Chat SDK brings agents to your users 360 billion tokens, 3 million customers, 6 engineers Meet the 2026 Vercel AI Accelerator Cohort Build knowledge agents without embeddings
Building efficient MCP servers - Vercel – Vercel
Andrew Qu · 2025-06-12 · via Vercel News

2 min read

The Model Context Protocol (MCP) standardizes how to build integrations for AI models. We built the MCP adapter to help developers create their own MCP servers using popular frameworks such as Next.js, Nuxt, and SvelteKit. Production apps like Zapier, Composio, Vapi, and Solana use the MCP adapter to deploy their own MCP servers on Vercel, and they've seen substantial growth in the past month.

MCP has been adopted by popular clients like Cursor, Claude, and Windsurf. These now support connecting to MCP servers and calling tools. Companies create their own MCP servers to make their tools available in the ecosystem.

The growing adoption of MCP shows its importance, but scaling MCP servers reveals limitations in the original design. Let's look at how the MCP specification has evolved, and how the MCP adapter can help.

Link to headingWhat's changed in MCP

When the first version of the MCP specification was published in November 2024, it supported Standard IO (stdio) and Server-Sent Events (SSE) as methods of transport between clients and servers. Stdio required running the MCP server locally, while SSE allowed servers to be hosted and accessed remotely.

However, SSE is an inefficient transport. It keeps a persistent connection open between client and server, even when idle. While Vercel's Fluid compute reduces overhead by reusing already allocated servers, SSE remains an unsustainable choice for MCP servers at scale.

Link to headingA new MCP server transport: Streamable HTTP

In March 2025, a new version of the MCP specification introduced Streamable HTTP as the recommended transport, replacing SSE. This change improves efficiency by removing the need for persistent client-server connections.

Existing investments and lack of resources have slowed the adoption of Streamable HTTP by both MCP client developers and MCP server developers. As a result, many existing MCP clients continue to support only stdio and SSE.

Server developers building with the Vercel MCP adapter get built-in support for both Streamable HTTP and SSE, with the option to disable SSE, to fully align with the officially supported and more efficient transport.

app/[transport]/route.ts

import { createMcpHandler } from '@vercel/mcp-adapter';

const handler = createMcpHandler(server => {

server.tool(

'roll_dice',

'Rolls an N-sided die',

{ sides: z.number().int().min(2) },

async ({ sides }) => {

const value = 1 + Math.floor(Math.random() * sides);

return { content: [{ type: 'text', text: `🎲 You rolled a ${value}!` }] };

}

);

});

export { handler as GET, handler as POST, handler as DELETE };

An example MCP server with a single tool call.

Link to headingOptimizing MCP servers for efficiency

Streamable HTTP, with its standard HTTP transport and support for both stateless and stateful server models, is the clear path forward for remote-hosted MCP servers.

To support clients that do not yet handle Streamable HTTP, the mcp-remote package can proxy it over stdio. With a small server setup change, like the approach used by Solana, developers can benefit from Streamable HTTP's efficiency without waiting for full client support.

This approach provides forward compatibility while client support continues to evolve. In most cases, mcp-remote will no longer be needed once native support for Streamable HTTP is more widely adopted.

One MCP server deployed on Vercel was able to cut over to Streamable HTTP completely and cut CPU usage in half, even with continued user growth.

Migrating to Streamable HTTP reduced CPU consumption by over 50%Migrating to Streamable HTTP reduced CPU consumption by over 50%

Migrating to Streamable HTTP reduced CPU consumption by over 50%

Link to headingBuilding for the future

The MCP ecosystem is evolving quickly. With Fluid compute and the MCP adapter, you can ship MCP servers that support both current and future clients.

We are committed to supporting every team adopting MCP. Explore the MCP Adapter and try the Next.js MCP template to build your MCP server.

MCP Server with Next.js

Get started building your first MCP server on Vercel.

Deploy now