DEV Community: curatedmcp

MCP Ecosystem Week 40: When Developer Demand Outpaces Your Allowlist

curatedmcp — Mon, 28 Sep 2026 10:54:40 +0000

MCP Ecosystem Week 40: When Developer Demand Outpaces Your Allowlist

The MCP ecosystem is consolidating around first-party integrations. This week's view counts tell a story: four of the five most-accessed servers are official platform integrations—GitHub, OpenAI, Figma, and Anthropic themselves. For platform teams, that's both a relief and a warning. Relief because first-party servers tend to ship with clearer auth models and audit trails. Warning because developer demand is moving faster than most allowlist policies can keep up.

This Week in MCP

No new servers entered the CuratedMCP catalog this week—a quiet moment in the review cycle, but the existing 79 risk-classified servers continue to see heavy adoption across Claude Code, Cursor, Windsurf, and GitHub Copilot. The pause in new reviews is a good time for platform teams to audit which of those 79 are actually running on your developer machines, versus which ones are sitting in your allowlist waiting for governance sign-off.

That distinction matters. Allowlist drift—where developers run tools that aren't formally approved—is a governance blind spot at most orgs rolling out AI coding agents. CuratedMCP's per-machine enforcement layer catches it, but only if you've turned on auditing.

On the Radar

The developer signal is unmistakable: GitHub Copilot MCP (98k views), OpenAI MCP (87k views), Figma MCP (82k views), and GitHub MCP (76k views) are the gravity wells in the ecosystem right now.

GitHub Copilot MCP and GitHub MCP need special care. GitHub Copilot MCP connects Copilot's code intelligence to any MCP client—powerful for completions and code review, but it extends your Copilot spend and audit surface across multiple IDEs and contexts. If you're already enforcing Copilot licensing per-seat, this server multiplies your compliance surface. GitHub MCP is the repo-and-workflow control plane—pull requests, issues, and actions all become accessible to AI agents running on developer machines. Before allowlisting, confirm: (1) your RBAC in GitHub is granular enough to prevent privilege escalation through the agent, and (2) your audit logs capture all agent-initiated actions.

OpenAI MCP surfaces GPT-4o, DALL-E, Whisper, and Embeddings. Check whether your org has a contracted relationship with OpenAI, and whether spinning up arbitrary Whisper jobs or embedding batches through an agent violates your spend controls or data residency policies.

Figma MCP is lower-risk from a data-access standpoint, but it's a supply-chain touchpoint—your design tokens, component libraries, and file structure become queryable by any LLM agent a developer runs. Vet it if design IP is sensitive.

Anthropic Claude MCP lets Claude call Claude—nested reasoning. It's elegant and low-risk from an auth perspective (same Anthropic API key), but watch for token-cost creep if developers use it for chained prompts without optimization discipline.

Governance Take

Here's the hard part: your developers are already using MCP servers you haven't approved yet. We see this across every customer. Cursor ships with bundled MCP servers, GitHub Copilot auto-enables integrations, and developers can point their IDEs at custom servers from a file path. Shadow MCP usage isn't hypothetical—it's happening now.

Start here: use CuratedMCP's auditor to scan what's actually running across your fleet, then compare it to your formal allowlist. You'll find gaps. Close them not by blocking servers, but by building a clear, low-friction approval workflow. If developers know they can get new servers audited and allowlisted in under a week, they're less likely to run unvetted ones.

Also: pair your allowlist with TokenShield. You'll get spend visibility across all your Claude usage (across Cursor, Code, Windsurf, Copilot), and an audit ledger of which agents and servers are burning tokens. That ledger is your governance evidence when budget holders ask why Claude spend doubled.

Govern MCP usage across your team with CuratedMCP — or scan your own stack free at https://www.curatedmcp.com/auditor.

Firebase MCP: Give Claude Full Access to Your Firebase Project

curatedmcp — Sun, 27 Sep 2026 10:54:30 +0000

Firebase MCP: Give Claude Full Access to Your Firebase Project

Firebase is everywhere in modern app development, but AI agents have historically been locked out. Firebase MCP changes that. Google's official Model Context Protocol server connects Claude, Cursor, Windsurf, and other AI tools directly to your Firebase project—giving them the ability to read and write Firestore documents, manage authentication, upload files to Storage, invoke Cloud Functions, and monitor your infrastructure in real time.

Instead of copying and pasting database schemas or wrestling with API documentation, you can now ask Claude to query your Firestore collections, create test users, debug function logs, or deploy new configuration—all within the conversation. The server uses service account credentials for secure, server-side access, so no personal auth tokens leak into your chat history.

What It Does

Firebase MCP exposes eight Firebase services as tools your AI agent can use:

Firestore: Full CRUD operations, complex queries, batch writes, and real-time listener setup
Authentication: Create users, assign roles, reset passwords, and audit sign-in events
Storage: Upload, download, list, and delete files programmatically
Cloud Functions: Invoke functions directly and stream their logs
Remote Config: Read and update feature flags without a deployment
Analytics: Query event data and user properties for debugging
Crashlytics: Access crash reports and see which users are affected
Hosting: Deploy static assets and manage redirects

This unlocks workflows that were previously manual: ask Claude to help you backfill user data in Firestore, debug authentication issues by reviewing sign-in logs, or quickly test a Cloud Function before pushing to production.

How to Install

npx -y firebase-mcp

Add it to your Claude Desktop config:

{
  "mcpServers": {
    "firebase-mcp": {
      "command": "npx -y firebase-mcp"
    }
  }
}

Authenticate with your Firebase service account key (download it from your Firebase console under Project Settings → Service Accounts).

Real-World Use Cases

Backfill user data: Ask Claude to migrate 500 old user records into Firestore, transform them to match your new schema, and report any validation errors—all in one go.
Debug auth issues: Have Claude pull recent sign-in audit logs, identify failed login patterns, and suggest fixes or reset passwords for locked accounts.
Test and monitor functions: Invoke a Cloud Function with test payloads, stream its logs in real time, and iterate on the code without manual deployments.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

Atlassian Remote MCP: Give Your AI Agent Access to Jira, Confluence, and Your Entire Work Graph

curatedmcp — Sat, 26 Sep 2026 10:54:29 +0000

Atlassian Remote MCP: Give Your AI Agent Access to Jira, Confluence, and Your Entire Work Graph

What It Does

Atlassian Remote MCP is the official bridge between AI agents (Claude, Cursor, Windsurf) and your Atlassian workspace. Instead of context-switching between your IDE and Jira/Confluence, your AI can now read issues, update sprint backlogs, search documentation, and create pages—all with OAuth-secured, permission-aware access.

The server unlocks two core capabilities:

Jira integration lets your agent create and search issues, run JQL queries for complex filtering, manage sprint backlogs, track epic progress, and transition work through your workflow. Works with both Jira Software and Service Management.

Confluence integration lets your agent search your knowledge base, read and create documentation, comment on pages, and navigate space hierarchies. Useful for pulling context from wiki pages or auto-generating documentation.

All actions respect your existing Atlassian permissions—your agent only sees what you'd see in Jira or Confluence.

How to Install

Install via npm:

npx -y @atlassian/mcp

Then add to your Claude Desktop config:

{
  "mcpServers": {
    "atlassian-remote-mcp": {
      "command": "npx -y @atlassian/mcp"
    }
  }
}

You'll authenticate via OAuth when you first connect. See the official Atlassian docs for setup on Cursor, Windsurf, and other clients.

Real-World Use Cases

Automate sprint planning: Ask Claude to search for all open bugs with a specific label, summarize blockers, and draft a sprint backlog update—all in one turn.
Pull context from your wiki: When reviewing code in Cursor, your agent can fetch relevant architecture docs or API standards from Confluence without you leaving the editor.
Generate issue summaries: Paste a GitHub PR link to Claude; it creates a Jira issue, links related epics, and updates the sprint board—no copy-paste required.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

Sentry MCP: Stop Copy-Pasting Stack Traces Into Claude

curatedmcp — Fri, 25 Sep 2026 10:54:29 +0000

Sentry MCP: Stop Copy-Pasting Stack Traces Into Claude

Error debugging just got faster. The official Sentry MCP server connects Claude, Cursor, and Windsurf directly to your Sentry project—pulling live error data, stack traces, and performance metrics without leaving your AI agent. No more screenshotting, no more manual context gathering. Ask Claude about an issue, and it has the full picture: what broke, who it affected, and what changed in the last deployment.

What It Does

Sentry MCP bridges the gap between error monitoring and AI-assisted debugging. When you ask Claude to help with a bug, it can now:

Query issues by project, environment, and time range — find the right error in seconds
Pull full stack traces with local variables — Claude sees exactly what your code had in memory when it crashed
Correlate errors with recent deployments — understand if a recent release introduced the problem
Access performance bottlenecks — get transaction data and slow query details
Search by user, URL, or tag — narrow scope to specific scenarios or affected customers
Suggest fixes with context — because Claude has breadcrumbs, release history, and affected user counts, not just a screenshot

The workflow is radically simpler: Claude sees an error → queries Sentry → gets full context → proposes a fix. No copy-paste friction. Built and maintained by Sentry, with OAuth-based auth so API keys never sit in plaintext configs.

How to Install

npx -y @sentry/sentry-mcp

Add to your Claude Desktop config:

{
  "mcpServers": {
    "sentry-mcp": {
      "command": "npx -y @sentry/sentry-mcp"
    }
  }
}

Authenticate via OAuth when you first use it. That's it.

Real-World Use Cases

"Claude, what caused the spike in 500 errors this morning?" — Point it at your Sentry project. Claude fetches issues from the last 2 hours, reads the stack traces, cross-references with your recent deploy, and tells you exactly which code change broke what.
"Help me fix this performance problem" — Claude pulls slow transaction data, sees which function is the bottleneck, and suggests optimizations with full context about user impact and affected endpoints.
"Which users hit this bug, and should I roll back?" — Claude queries affected users, reads breadcrumbs to understand their actions, checks release health metrics, and gives you a rollback recommendation backed by actual error data.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

MCP Ecosystem Week 39: Five New Servers, One Governance Question — Who Controls the Integration Layer?

curatedmcp — Mon, 21 Sep 2026 10:54:40 +0000

MCP Ecosystem Week 39: Five New Servers, One Governance Question — Who Controls the Integration Layer?

The MCP marketplace added five new risk-classified servers this week, and they tell a story platform teams need to hear. Three of them—send21, Kleap, and ReefAPI MCP—hand AI agents direct access to external systems: payments, web publishing, and live API data. Two others—ContextStream and Robot Speed—operate inside your perimeter but expand what agents can see and do. All are free, all are now risk-classified, and all demand a clear allowlist decision before a single developer uses them.

This Week in MCP

send21 lets agents draft payments; humans sign in their own wallet. Non-custodial, which is good—your agents can't move money without explicit approval. But allowlist with care: define which teams use it, audit transaction drafts, and ensure signers understand what they're approving.

ContextStream centralizes project decisions and learned context for AI agents—a shared memory layer. Platform teams should ask: who owns this context repository? Is it version-controlled? Can it leak proprietary lessons or architecture decisions to new agents or team members?

Kleap gives agents the ability to create and publish live websites. This is a direct supply-chain risk: a compromised agent or malicious prompt injection could publish phishing sites or malware distribution points under your domain. Require explicit approval per developer, tight scope, and audit logs.

Robot Speed is SEO and traffic analysis—low-risk from a data access perspective, but consider: are agents learning competitive intelligence from your own sites? Monitor for accidental leaks.

ReefAPI MCP exposes 250+ live web APIs (e-commerce, jobs, finance, social). This is broad external data access. Allowlist only if your team has a clear use case, understands rate limits, and can audit API calls in your ledger.

On the Radar

The five most-viewed servers this week are all official integrations or nesting patterns: GitHub Copilot MCP, OpenAI MCP, Figma MCP, GitHub MCP, and Anthropic Claude MCP. High demand makes sense—they're foundational to AI coding workflows. But governance friction is real.

GitHub Copilot MCP and GitHub MCP both bridge GitHub; teams may allowlist one without realizing they've approved repo and workflow access twice. OpenAI MCP and Anthropic Claude MCP create sub-agent chains—elegant for specialization, opaque for audit. Figma integration is low-risk but ties design-to-code velocity to third-party API availability.

For each, platform teams should verify: Does your SSO cover the MCP server's auth model, or do developers authenticate directly? Are API credentials scoped per user or shared? Can you revoke access atomically across all IDEs?

Governance Take

Here's the hard truth: most platform teams can see which MCP servers developers install in Claude, but they're flying blind across Cursor, Windsurf, and local IDE plugins. Allowlisting one server in one tool doesn't stop a developer from running the same server elsewhere. That's allowlist drift, and it's your biggest governance gap right now.

Add spend visibility on top, and the picture gets messier. TokenShield gives you a live ledger of Claude spend and an opt-in optimization layer—visibility first, then measured efficiency. But if you can't see which servers are running, you can't tie spend to integration choices. You can't tell your CFO why one team's token bill spiked. You can't audit what data left your network.

This week's five new servers—especially send21, Kleap, and ReefAPI—are all high-impact integrations. Before you allowlist any of them, you need per-machine enforcement across all your AI coding tools, a unified audit log, and a clear policy about external data access.

Govern MCP usage across your team with CuratedMCP — or scan your own stack free at https://www.curatedmcp.com/auditor.

MCP Ecosystem Week 38: Why Your MCP Allowlist Matters More Than Your Server Count

curatedmcp — Mon, 14 Sep 2026 10:39:43 +0000

MCP Ecosystem Week 38: Why Your MCP Allowlist Matters More Than Your Server Count

The MCP catalog has hit 72 risk-classified servers, but this week's signal isn't about volume—it's about velocity. The most-viewed servers are integrations with GitHub, OpenAI, Figma, and Claude itself, which means your developers are already asking for deep connections between their coding tools and external services. If you haven't formally decided whether to allowlist these, you're operating blind. This week we break down what platform teams need to know.

This Week in MCP

No new servers entered the catalog this week, but that's not a slowdown—it's stabilization. The existing 72 risk-classified servers now represent the most common ask patterns: GitHub and cloud IDE workflows, design tooling, foundational model APIs, and internal tool connectors.

The real story is which servers your team is already using. The view counts this week show heavy demand for official integrations (OpenAI, Anthropic, GitHub, Figma). If you see developers spinning up MCP servers without your approval, they're almost certainly reaching for these exact tools. A governance-first approach means:

Review before deployment, not after. Run these through your risk framework now, before they appear in your audit logs as shadow tools.
Document your decision. Whether you allowlist or block, a written policy prevents re-litigating the same request three times.
Version-lock approved servers. Developers will install the latest version; you need visibility into which versions run on which machines.

On the Radar

The five most-viewed servers this week represent three governance tiers:

Tier 1: Native IDE integrations — GitHub Copilot MCP and Anthropic Claude MCP sit at the top because they're native to the workflows developers already use. Governance question: does nesting Claude within Claude create audit-log complexity? Yes—plan for deeper tracing if you allowlist this.

Tier 2: High-privilege external APIs — OpenAI MCP and GitHub MCP both grant access to production data (code, repos, PRs). Before allowlisting, confirm: Is auth scoped to what developers actually need? Can you audit which repos/models each developer accesses? GitHub MCP in particular bypasses the GitHub web UI's audit trail—you'll need MCP logs to see who touched which PR.

Tier 3: Design and structured data — Figma MCP is lower-risk on the surface (design files aren't secrets), but consider: does your Figma workspace contain proprietary design systems or strategic mockups? What's the auth story if a developer leaves?

Governance Take

Here's what we're seeing across platform teams this week: allowlist drift. A Head of Platform approves five MCP servers across Claude Code and Cursor. Two weeks later, developers are running eight servers across Windsurf and GitHub Copilot, with two unapproved. Why? IDE-specific tooling, unclear policy inheritance, and no machine-level enforcement.

The fix isn't a longer spreadsheet—it's per-machine enforcement. CuratedMCP lets you define one allowlist and push it to every developer's machine across every IDE. You also get continuous visibility: TokenShield logs every MCP call, so if an unapproved server spins up, you see it immediately. That's not just a cost-control tool; it's your audit layer.

One more thing: token spend sprawl follows the same pattern as MCP sprawl. Developers add unapproved servers, those servers call Claude or OpenAI models, and your bill grows without anyone knowing why. TokenShield gives you spend visibility plus the ability to see which MCP server triggered which API call—so you can tie governance decisions to actual cost.

Start here: scan your team's current MCP usage this week. You'll probably find more servers running than you've approved.

Govern MCP usage across your team with CuratedMCP — or scan your own stack free at https://www.curatedmcp.com/auditor.

MCP Ecosystem Week 37: The Governance Challenge of Nested AI and Cross-IDE Control

curatedmcp — Mon, 07 Sep 2026 10:39:38 +0000

MCP Ecosystem Week 37: The Governance Challenge of Nested AI and Cross-IDE Control

As AI coding tools proliferate across your engineering org, a critical governance gap is emerging: developers are spinning up MCP servers faster than platform teams can audit them. Week 37 reveals which integrations your team is already using—and what security decisions you need to make about them.

This Week in MCP

No new servers entered the CuratedMCP catalog this week, but that's not a slowdown—it's a sign of maturity. The existing 72 risk-classified servers are seeing accelerating adoption. This is the moment to ensure your allowlist policies are enforced consistently across Claude Code, Cursor, Windsurf, and GitHub Copilot. If you haven't conducted a live audit of which servers are running on developer machines right now, Week 37 is your reminder: shadow MCP usage is real, and it's growing.

On the Radar

The five most-viewed servers this week tell you exactly what your developers want to connect to their AI agents:

GitHub Copilot MCP (98K views) and GitHub MCP (76K views) dominate. Both grant AI agents deep access to your source code, pull requests, and workflows. Governance consideration: Does your team require mutual TLS or IP allowlisting for these integrations? Are audit logs being retained?

OpenAI MCP (87K views) lets developers route requests through OpenAI's APIs from within Claude or other clients. Supply-chain risk: You're now dependent on two LLM vendors' uptime and data handling policies. Token spend visibility becomes essential when requests can flow through multiple providers.

Figma MCP (82K views) exposes design tokens and component libraries to AI agents. AppSec angle: Who owns the Figma workspace credentials being stored locally? Is there a credential rotation policy?

Anthropic Claude MCP (76K views) allows nesting Claude as a sub-agent. This is architectural—it adds latency, multiplies token consumption, and creates audit visibility challenges across multiple inference calls.

Governance Take

Here's what keeps platform leaders awake: you cannot govern what you cannot see. This week's view counts reflect organic developer demand, but they likely undercount actual deployments. Developers are running MCP servers across four separate IDE integrations (Claude Code, Cursor, Windsurf, GitHub Copilot), each with its own local configuration. Your RBAC and allowlist are only as strong as your ability to enforce them per-machine.

Start here: audit which MCP servers are actually running in your org right now. Use CuratedMCP's per-machine enforcement to lock down policy—don't rely on developer best practices. Then, layer in visibility: when developers connect OpenAI MCP or Figma MCP, you should know about it the same way you monitor SaaS app usage. If you're seeing multiple Claude integrations and nested Claude calls, you're also seeing token spend sprawl. TokenShield gives you the visibility to measure that spend and understand which optimization patterns actually work in your environment—before you commit to cost-saving approaches that might degrade your team's velocity.

The policy library at 72 risk-classified servers means you have a foundation. Now enforce it.

Govern MCP usage across your team with CuratedMCP — or scan your own stack free at https://www.curatedmcp.com/auditor.

AWS MCP Suite: Query and Manage Your Entire AWS Cloud via Natural Language

curatedmcp — Mon, 07 Sep 2026 10:39:30 +0000

AWS MCP Suite: Query and Manage Your Entire AWS Cloud via Natural Language

What It Does

The AWS MCP Suite is the official bridge between AI agents and your AWS infrastructure. Instead of wrestling with SDK boilerplate, IAM policy syntax, and credential management, you can now ask Claude or Cursor to inspect EC2 instances, query DynamoDB tables, upload to S3, invoke Lambda functions, and manage RDS databases—all through conversation.

The suite wraps seven core AWS services (DynamoDB, S3, Lambda, EC2, RDS/Aurora, CloudWatch, and IAM) into a permission-aware protocol. Your agent inherits your existing AWS credentials—IAM roles, profiles, or environment variables—so there's no extra auth layer to configure. Critically, scoped permissions mean your agent only touches what you've explicitly allowed via your IAM setup. You stay in control.

Real wins: You can ask your agent to "list all EC2 instances in us-east-1 and stop the ones tagged 'dev'," or "create a snapshot of my production RDS database and show me the latest CloudWatch metrics," or "upload this CSV to S3 and update the bucket lifecycle policy." The agent handles the API calls, error handling, and response parsing. You get structured, natural-language infrastructure management.

How to Install

npx -y @aws/mcp

Add it to your Claude Desktop config:

{
  "mcpServers": {
    "aws-mcp-suite": {
      "command": "npx -y @aws/mcp"
    }
  }
}

Same process for Cursor or Windsurf. Your AWS credentials are picked up automatically from ~/.aws/credentials, environment variables, or IAM roles (if running on EC2/Lambda).

Real-World Use Cases

Infrastructure auditing: Ask Claude to scan all your EC2 security groups and report which ones allow unrestricted inbound traffic on port 22, then remediate them in one shot.
Database operations: Query RDS logs, create snapshots, and check Aurora performance metrics without leaving your editor—great for debugging production issues on the fly.
CI/CD helpers: Invoke Lambda functions directly from Cursor to trigger deployments, check job status, or pull CloudWatch logs—no terminal tab-switching needed.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

Linear MCP: Stop Context-Switching Between Claude and Linear

curatedmcp — Sun, 06 Sep 2026 10:39:34 +0000

Linear MCP: Stop Context-Switching Between Claude and Linear

Engineering teams switch between their AI agent and Linear dozens of times a day. You ask Claude for help debugging, then manually hop over to Linear to create a ticket. You're in a sprint planning call, need to check capacity, and alt-tab away from your IDE. Linear MCP eliminates that friction by giving your AI agent direct access to your entire issue tracker.

This is the official Linear integration—built and maintained by Linear themselves. It connects Claude, Cursor, Windsurf, and other MCP-compatible agents to your Linear workspace with full OAuth support, so each team member's permissions stay intact. Your AI can now create issues, manage sprints, update statuses, query workload, and pull metrics without you leaving your flow.

What It Does

Linear MCP unlocks five core capabilities:

Issue Management: Create fully-formed bugs and features with title, description, assignee, labels, and priority in a single prompt. Search and filter across projects, assignees, statuses, or labels to find exactly what you need.

Workflow Automation: Move issues through your team's workflow states, add comments with context (like PR links), and link related tickets—all without touching the Linear UI.

Sprint & Cycle Visibility: Query what's in the current sprint, check team capacity, and get cycle completion metrics to answer "what did we ship last week?" in seconds.

Project Strategy: Access your roadmap, explore project hierarchy, and check milestone status for planning conversations with your AI agent.

Per-User Security: OAuth authentication means each team member's access is scoped to what Linear already allows them to see.

The result: your AI agent becomes a genuine productivity multiplier for engineering teams, not just a chatbot you paste code into.

How to Install

Install via npm:

npx -y @linear/mcp

Then add Linear MCP to your Claude Desktop configuration:

{
  "mcpServers": {
    "linear-mcp": {
      "command": "npx -y @linear/mcp"
    }
  }
}

Authenticate with your Linear workspace when prompted, and you're live.

Real-World Use Cases

Instant Bug Reports: Say "Create a high-priority bug: login timeout on mobile, affects production" and Claude generates the ticket with full details and assigns it to you—no manual typing.
Sprint Triage in Chat: Ask Claude "Show me all blocked issues in the current sprint" and get a filtered list with context, ready to discuss in standup.
Link PR to Issue: After code review, tell Claude "Move ENG-431 to In Review and add a comment with [PR link]"—one command, no context switch.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

Vercel MCP: Debug Production Errors Without Leaving Your Editor

curatedmcp — Sat, 05 Sep 2026 10:39:32 +0000

Vercel MCP: Debug Production Errors Without Leaving Your Editor

The worst part of Next.js development isn't writing code—it's the "works locally but not on Vercel" nightmare. You deploy, something breaks in production, and you're stuck copy-pasting logs into ChatGPT while Claude waits helplessly outside your deployment dashboard. Vercel MCP fixes that.

This official Vercel integration brings your entire deployment workflow directly into Claude, Cursor, or Windsurf. Your AI agent can now inspect deployments, pull build logs from failing builds, manage environment variables, check edge function analytics, and even create new projects—all without you manually hunting through the Vercel dashboard.

What It Does

Vercel MCP unlocks these capabilities for your AI agent:

List and inspect deployments across all your projects with real-time status
Pull build logs from failing deployments so Claude can actually see what went wrong
Manage environment variables (create, update, delete) for any deployment environment
Create and configure projects from scratch when you need to spin up new ones
Monitor deployment health including build times and performance metrics
Access edge function logs and analytics for debugging serverless issues
Handle custom domains and DNS configuration programmatically
Manage team members and permissions across your Vercel organization

The key difference: instead of describing errors in words, Claude reads the actual logs. It traces the stack trace, identifies the root cause, and suggests a fix based on what actually happened in your build.

How to Install

Install the Vercel MCP server with a single command:

npx -y @vercel/mcp

Then add it to your Claude Desktop config:

{
  "mcpServers": {
    "vercel-mcp": {
      "command": "npx -y @vercel/mcp"
    }
  }
}

You'll need a Vercel API token set as the VERCEL_API_TOKEN environment variable.

Real-World Use Cases

Debug a failed production deployment: Paste your project URL and ask Claude to fetch the latest build logs, identify the error, and suggest a fix—all in one turn.
Sync environment variables across environments: Ask Claude to read your production env vars and replicate them to staging, ensuring consistency without manual copy-paste.
Investigate DNS or domain issues: Have Claude check your custom domain configuration, DNS records, and SSL certificate status when deployments aren't accessible.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

Figma MCP: Read Design Tokens and Components Directly Into Your AI Coding Session

curatedmcp — Fri, 04 Sep 2026 10:39:35 +0000

Figma MCP: Read Design Tokens and Components Directly Into Your AI Coding Session

Design handoff friction is real. You're building a component, context-switch to Figma to grab a hex value, squint at the typography panel, guess at spacing. Figma MCP eliminates that loop entirely.

This official Figma integration lets Claude, Cursor, and Windsurf read your design files directly—no screenshots, no manual specs. Your AI agent can extract exact token values, component definitions, Auto Layout constraints, and frame measurements. It's the design-to-code gap closed.

What It Does

Figma MCP gives AI agents programmatic access to your Figma workspace. Read file contents, component variants, and design tokens in real time. Extract precise hex colors, font weights, line heights, spacing scales, border radius values, and responsive layout constraints without leaving your editor.

The server handles the structural stuff too: navigate page hierarchies, read frame coordinates and dimensions, access component properties, pull frame comments and annotations. You can even export specific frames as images for reference.

The impact is immediate. Before: "Make the button padding match the design." After: AI reads padding: 12px, background: #2563EB, font-weight: 600 directly from your Figma file and implements it. Pixel-perfect, first try.

How to Install

Add Figma MCP to your local environment:

npx -y figma-developer-mcp

Configure it in Claude Desktop (or Cursor/Windsurf—docs linked below):

{
  "mcpServers": {
    "figma-mcp": {
      "command": "npx -y figma-developer-mcp"
    }
  }
}

You'll need a Figma personal access token in your environment. See the docs for setup details.

Real-World Use Cases

Token-driven component building: Ask Claude to build a button component and specify "use tokens from my Figma file." AI reads your exact color palette, typography scale, and spacing tokens, then generates components that match your design system without manual copying.
Design system documentation: Extract all component variants, their properties, and constraints in one go. Feed this to Claude to auto-generate component stubs, prop types, or Storybook documentation.
Responsive layout implementation: Have Claude read your Auto Layout rules and responsive breakpoints directly from Figma frames, then generate CSS or React layouts that respect your design constraints.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

Cloudflare MCP: Manage Your Edge Infrastructure Through AI

curatedmcp — Thu, 03 Sep 2026 10:39:32 +0000

Cloudflare MCP: Manage Your Edge Infrastructure Through AI

Cloudflare MCP is the official integration that lets Claude, Cursor, and other AI agents control your entire Cloudflare ecosystem via natural language. No manual dashboard clicks. No copy-pasting API calls. Just tell your AI what you need, and it deploys Workers, manages databases, updates DNS records, and monitors analytics in real time.

What It Does

This server bridges AI agents directly to Cloudflare's infrastructure layer. You get full programmatic access to:

Cloudflare Workers — deploy, update, and manage serverless functions
KV & D1 — read/write configuration data and query SQLite databases
R2 Object Storage — upload, list, and manage files at edge scale
DNS & Pages — configure zones, records, and static site deployments
Analytics & WAF — monitor traffic patterns and security rules

The key unlock: instead of context-switching between your IDE, the Cloudflare dashboard, and API docs, you describe what you want in plain English. "Deploy a new Worker that handles CORS preflight requests" or "Show me my zone's request volume for the last 24 hours"—and it just works.

Authentication uses OAuth, so your credentials stay secure and no API keys are stored locally.

How to Install

npx -y @cloudflare/mcp-server-cloudflare

Add to your Claude Desktop config:

{
  "mcpServers": {
    "cloudflare-mcp": {
      "command": "npx -y @cloudflare/mcp-server-cloudflare"
    }
  }
}

Then restart Claude and authenticate via the OAuth flow when prompted.

Real-World Use Cases

Deploy a CORS-handling Worker on demand — "Create a new Worker that proxies requests to my API and adds the correct CORS headers for my domain"
Automate configuration storage — "Store this feature flag configuration in KV namespace my-config with a 24-hour TTL"
Debug infrastructure at scale — "Show me analytics for the last 7 days, broken down by country, and identify any traffic spikes"

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.