--- title: MCP standardized tools. Now we need the infrastructure... description: MCP (Model Context Protocol) solved the problem of standardizing how AI models talk to tools, with adoption moving extremely fast across Claude Code, Cursor,... canonical: https://daily.dev/posts/mcp-standardized-tools-now-we-need-the-infrastructure-around-them--qv8rzplng twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: MCP standardized tools. Now we need the infrastructure around them. | daily.dev og:description: MCP (Model Context Protocol) solved the problem of standardizing how AI models talk to tools, with adoption moving extremely fast across Claude Code, Cursor,... og:url: https://daily.dev/posts/mcp-standardized-tools-now-we-need-the-infrastructure-around-them--qv8rzplng og:image: https://api.daily.dev/og/posts/Qv8RZplNg.png og:image:alt: MCP standardized tools. Now we need the infrastructure around them. og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # MCP standardized tools. Now we need the infrastructure around them. **[Stack Overflow Blog](https://daily.dev/sources/stackov)** · 10 min read · 0 upvotes · 0 comments ## Summary MCP (Model Context Protocol) solved the problem of standardizing how AI models talk to tools, with adoption moving extremely fast across Claude Code, Cursor, Codex, and VS Code. But the author argues the protocol's success has exposed a bigger gap: the operational infrastructure around it remains immature. Teams repeatedly rebuild authentication, deployment, retries, logging, secret handling, and rate limiting from scratch. Local stdio servers worked well for experimentation but break down at team scale, while remote servers introduce authentication, multi-tenancy, and latency challenges. Context management is becoming its own infrastructure problem, with progressive tool discovery emerging to avoid flooding models with irrelevant tool schemas. Security risks escalate as agents gain autonomy, citing a 2026 study finding 40.55% of 7,973 live remote MCP servers exposed tools without authentication. The piece draws parallels to HTTP, which required CDNs, load balancers, and identity systems to become production-grade, arguing MCP infrastructure may become its own distinct layer. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What percentage of live remote MCP servers were found to expose tools without authentication? A 2026 measurement study of 7,973 live remote MCP servers found that 40.55% exposed tools without any authentication. This is described as a deployment problem rather than a flaw in the Model Context Protocol itself, but it represents a real risk teams encounter when running remote MCP servers in production. _Teams hardening MCP deployments can track security research like this through daily.dev._ ### What changed in the July 2026 MCP specification update? The July 2026 Model Context Protocol specification update removed protocol-level sessions and the initialize handshake, letting remote MCP servers scale more like ordinary HTTP services. This helps the remote deployment case specifically, though it does not remove the need for authentication, authorization, or broader operational discipline around running servers reliably. _Developers tracking MCP spec changes can follow protocol updates like this on daily.dev._ ### Why does giving an AI agent too many MCP tools hurt performance? Giving an agent 50, 100, or several hundred tools can reduce tool-selection accuracy because the context fills with schemas the model may never use, and name collisions or vague descriptions make selection worse. Progressive or dynamic tool discovery, which defers full tool definitions and uses a search-style meta-tool, is being recommended to keep models focused on a smaller relevant set of tools. _Anyone designing agent tool integrations can follow context engineering patterns like this on daily.dev._ ## Similar posts on daily.dev - [MCP is a fad](https://daily.dev/posts/mcp-is-a-fad-yboi44ud0) · Hacker News · 2 upvotes · 0 comments - [Why the Model Context Protocol Won](https://daily.dev/posts/why-the-model-context-protocol-won-sebuwojje) · The New Stack · 3 upvotes · 0 comments - [Model Context Protocol Explained in 3 Levels of Difficulty](https://daily.dev/posts/model-context-protocol-explained-in-3-levels-of-difficulty-qr2eqsxue) · Machine Learning Mastery · 1 upvotes · 0 comments - [The role of MCP in context engineering](https://daily.dev/posts/the-role-of-mcp-in-context-engineering-xktcrwru5) · InfoWorld · 2 upvotes · 0 comments - [MCP for Technical Professionals: A Comprehensive Guide to Understanding and Implementing the Model Context Protocol](https://daily.dev/posts/mcp-for-technical-professionals-a-comprehensive-guide-to-understanding-and-implementing-the-model-c-xnb8mnykd) · Security Boulevard · 1 upvotes · 0 comments --- Tags: [#security](https://daily.dev/tags/security), [#ai-agents](https://daily.dev/tags/ai-agents), [#observability](https://daily.dev/tags/observability), [#mcp](https://daily.dev/tags/mcp), [#context-engineering](https://daily.dev/tags/context-engineering) [View this post on daily.dev](https://daily.dev/posts/mcp-standardized-tools-now-we-need-the-infrastructure-around-them--qv8rzplng) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"MCP standardized tools. Now we need the infrastructure around them.","url":"https://daily.dev/posts/mcp-standardized-tools-now-we-need-the-infrastructure-around-them--qv8rzplng","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/mcp-standardized-tools-now-we-need-the-infrastructure-around-them--qv8rzplng"},"datePublished":"2026-10-16T05:59:15.775Z","dateModified":"2026-10-16T05:59:36.728Z","description":"MCP (Model Context Protocol) solved the problem of standardizing how AI models talk to tools, with adoption moving extremely fast across Claude Code, Cursor,...","image":"https://media.daily.dev/image/upload/s--foaA6JGU--/f_auto/v1722860399/public/Placeholder%2004","thumbnailUrl":"https://media.daily.dev/image/upload/s--foaA6JGU--/f_auto/v1722860399/public/Placeholder%2004","isAccessibleForFree":true,"articleSection":"Stack Overflow Blog","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Stack Overflow Blog","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/003e14873238469caea60ebdd34508ae","url":"https://daily.dev/sources/stackov"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/mcp-standardized-tools-now-we-need-the-infrastructure-around-them--qv8rzplng","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"security,ai-agents,observability,mcp,context-engineering","timeRequired":"PT10M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Stack Overflow Blog","item":"https://daily.dev/sources/stackov"},{"@type":"ListItem","position":3,"name":"MCP standardized tools. Now we need the infrastructure around them."}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/mcp-standardized-tools-now-we-need-the-infrastructure-around-them--qv8rzplng#faq","mainEntity":[{"@type":"Question","name":"What percentage of live remote MCP servers were found to expose tools without authentication?","acceptedAnswer":{"@type":"Answer","text":"A 2026 measurement study of 7,973 live remote MCP servers found that 40.55% exposed tools without any authentication. This is described as a deployment problem rather than a flaw in the Model Context Protocol itself, but it represents a real risk teams encounter when running remote MCP servers in production. Teams hardening MCP deployments can track security research like this through daily.dev."}},{"@type":"Question","name":"What changed in the July 2026 MCP specification update?","acceptedAnswer":{"@type":"Answer","text":"The July 2026 Model Context Protocol specification update removed protocol-level sessions and the initialize handshake, letting remote MCP servers scale more like ordinary HTTP services. This helps the remote deployment case specifically, though it does not remove the need for authentication, authorization, or broader operational discipline around running servers reliably. Developers tracking MCP spec changes can follow protocol updates like this on daily.dev."}},{"@type":"Question","name":"Why does giving an AI agent too many MCP tools hurt performance?","acceptedAnswer":{"@type":"Answer","text":"Giving an agent 50, 100, or several hundred tools can reduce tool-selection accuracy because the context fills with schemas the model may never use, and name collisions or vague descriptions make selection worse. Progressive or dynamic tool discovery, which defers full tool definitions and uses a search-style meta-tool, is being recommended to keep models focused on a smaller relevant set of tools. Anyone designing agent tool integrations can follow context engineering patterns like this on daily.dev."}}]} ```