Best of MCP — April 2026

  1. 1
    Article
    Avatar of addyAddy Osmani·6w

    Agentic Engine Optimization (AEO)

    Agentic Engine Optimization (AEO) is a new discipline for structuring technical documentation so AI coding agents can effectively discover, parse, and use it. Unlike human readers, agents like Claude Code, Cursor, and Cline compress multi-page navigation into single HTTP requests, bypass all client-side analytics, and silently discard content that exceeds their context windows. Key AEO practices include: auditing robots.txt to allow AI agent traffic, adding an llms.txt sitemap for agent discovery, writing skill.md files that declaratively describe API capabilities, keeping documentation pages under token limits (15K–25K tokens for most pages), serving clean Markdown alongside HTML, surfacing token counts as metadata, adding 'Copy for AI' buttons, and creating AGENTS.md files in repositories. A companion open-source audit tool called agentic-seo automates checking for these signals.

  2. 2
    Article
    Avatar of thoughbotthoughbot·6w

    The AI stack I developed to help me context switch

    A product manager shares a personal AI-powered workflow for managing context switching across multiple projects. The system combines three tools: Gemini Meeting Notes for automatic meeting capture, Obsidian as a local markdown knowledge base, and Claude Projects for AI-assisted context retrieval. A key element is a master index file called HIVEMIND.md that instructs Claude to load only relevant project files, keeping sessions lean. An MCP plugin connects Obsidian to Claude Desktop, while a daily automated job pulls Gemini notes into Obsidian. The advice is to adopt the stack incrementally, starting with whichever problem is most pressing.

  3. 3
    Article
    Avatar of collectionsCollections·6w

    Claude Code now supports automated routines triggered by schedule, API, or GitHub events

    Anthropic has launched Routines in Claude Code, a cloud-based automation feature currently in research preview for Pro, Max, Team, and Enterprise subscribers. Routines let you configure a task once — with a prompt, repo, and connectors — and have it run automatically on Anthropic's infrastructure without a local machine. Three trigger types are supported: scheduled (hourly/daily/weekly), API (HTTP POST with bearer token), and GitHub events (PRs, pushes, releases, issues). Multiple triggers can be combined on a single routine. Routines run fully autonomously with access to MCP servers, GitHub repos, and environment variables. Daily run limits vary by plan (5 for Pro, 15–25 for Max, 25 for Team/Enterprise) and run against subscription billing rather than API billing. Key caveats include LLM inference latency (2–15s per step), no pre-built OAuth connectors, and the need for careful prompt scoping since routines run unattended. Anthropic positions this as a developer-native alternative to tools like Make.com and n8n, with plain prompts and Git integration instead of a visual node editor.

  4. 4
    Article
    Avatar of iwzfqwgzjuz3tmf4zw9azAnmol Baranwal·4w

    Generative UI explained without the hype

    Generative UI is a spectrum of three patterns for how AI agents control UI: Controlled (agent picks from predefined components), Declarative/A2UI (agent selects from a schema-driven catalog), and Open-ended (agent generates raw HTML or controls external apps via MCP). Each pattern trades design control for flexibility. CopilotKit supports all three via the AG-UI protocol, which streams events between agent and frontend. The post demystifies the vague term and explains when each pattern is appropriate, with code examples for each approach.

  5. 5
    Article
    Avatar of phProduct Hunt·8w

    Baton: Orchestrate your AI coding agents

    Baton is a desktop app for orchestrating multiple AI coding agents in parallel. Each agent runs in its own git-isolated workspace, with smart notification badges to flag which agents need attention. It supports Claude Code, Codex, OpenCode, and any terminal-based agent. Features include diff review, file browsing, codebase search, and a built-in MCP server that lets agents spawn new agents. Built by a developer who needed a single unified interface to manage multiple agents without constant window-switching.

  6. 6
    Article
    Avatar of supabaseSupabase·7w

    AI Agents Know About Supabase. They Don't Always Use It Right.

    Supabase has released Agent Skills, an open-source set of instructions that teach AI coding agents how to use Supabase correctly. The problem: agents already know about Supabase from training data, but they make critical mistakes — skipping RLS policies, hallucinating CLI commands, creating views without security_invoker=true, and ignoring up-to-date docs. The skill is a ~100-line SKILL.md file covering four areas: documentation access (teaching agents to fetch current docs via MCP, curl, or web search), security (inline checklist of RLS and auth gotchas), tooling (CLI --help discovery and MCP troubleshooting), and schema management (direct SQL edits during dev, then formalize migrations). Testing across Claude Code and Codex showed consistent improvement when using MCP + Skill vs. baseline or MCP alone, with task completion rates rising from 42-71% to 67-88%. Key insight: the bottleneck is context, not capability — agents knew how to implement security_invoker correctly when the skill was loaded, they just didn't know when to apply it.

  7. 7
    Article
    Avatar of flutterFlutter·8w

    How Dart and Flutter are thinking about AI in 2026

    The Flutter and Dart team shares their strategic thinking on AI integration for 2026, addressing three developer personas: traditional developers, AI-assisted developers, and AI-first developers. Key data points include 79% of Flutter developers using AI assistants and a 46% trust gap in AI accuracy. The team outlines guiding principles including 'humans first' language design, agent-agnostic open standards via MCP, and reducing the 'verification tax' through quality AI-generated code. Collaborations with Google DeepMind and Antigravity are mentioned, along with MCP tooling for hot reload during AI-assisted development.

  8. 8
    Article
    Avatar of iwzfqwgzjuz3tmf4zw9azAnmol Baranwal·5w

    8 Generative UI projects worth cloning this weekend

    A curated list of 8 open-source Generative UI projects worth exploring, sparked by a reverse-engineered version of Claude's GenUI. Projects range from a native macOS terminal agent with streaming widgets, an AI-powered ShadCN component generator, an Excalidraw diagram chat loop, a voice agent with mid-call form injection, to an MCP-based lazy-loading UI guidelines system. All are free, open source, and include GitHub links with live demos where available.

  9. 9
    Article
    Avatar of cloudflareCloudflare·6w

    Register domains wherever you build: Cloudflare Registrar API now in beta

    Cloudflare has launched a beta Registrar API that lets developers and AI agents programmatically search for domains, check real-time availability, and register them at cost — all without leaving their editor, terminal, or agentic workflow. The API exposes three core endpoints: search (cached results), check (authoritative registry query), and register (completes synchronously in seconds or returns a 202 for polling). WHOIS privacy is enabled by default at no extra charge, and the API integrates with the Cloudflare MCP, making it immediately accessible from tools like Cursor and Claude Code. At-cost pricing applies regardless of whether registration happens via dashboard, API, or agent. Future plans include lifecycle management features such as transfers, renewals, and contact updates, plus a broader registrar-as-a-service offering.

  10. 10
    Article
    Avatar of thomasthorntonThomas Thornton·5w

    Terraform Module Design Is the Hard Part – So I Built an Agent Skill for It

    Building Terraform modules is easy; designing them well is not. This post introduces a GitHub Copilot agent skill called terraform-module-creator that prioritizes design judgment over code generation. Rather than turning a prompt directly into files, the skill follows a deliberate workflow: evaluating whether a module should exist, defining its boundary, designing a minimal interface, and only then generating code. It integrates Azure MCP and HashiCorp Terraform Registry MCP to pull live provider documentation and best practices at design time. The skill produces full module scaffolding including README, terraform-docs config, runnable examples, and validation steps (fmt, validate, tflint). It also supports review and refactor modes for existing modules, and encourages versioning discipline for shared modules. Core design principles include KISS, DRY with judgment, clear responsibility, and opinionated safe defaults.

  11. 11
    Article
    Avatar of wordpressdevWordPress Developer·7w

    What’s new for developers? (April 2026)

    WordPress 7.0's release cycle has been extended to fix a performance issue in the Real-Time Collaboration (RTC) database architecture. Pre-release versions are paused until April 17th, with a new schedule announced by April 22nd. Key developer highlights include: RTC using Yjs/CRDT with HTTP polling, requiring migration from classic meta boxes; a new WP AI Client PHP library providing a unified abstraction for AI providers (OpenAI, Anthropic, Google, Ollama, Mistral, OpenRouter); a Connectors API for platform-level AI credential management; a Client-Side Abilities API in JavaScript; and WordPress Playground's new MCP server enabling AI coding agents like Claude Code to control a local WordPress instance via WebSocket. Theme developers get pseudo-state button styling in theme.json, viewport-based block visibility controls, and background gradient support. PHP minimum requirement rises to 7.4, with 8.2+ recommended.

  12. 12
    Article
    Avatar of cloudflareCloudflare·5w

    The AI engineering stack we built internally — on the platform we ship

    Cloudflare shares how they built their internal AI engineering stack using their own products over 11 months. The stack includes Cloudflare Access for zero-trust auth, AI Gateway for centralized LLM routing (20M+ requests/month, 241B tokens), Workers AI for on-platform inference, MCP Server Portals aggregating 13 servers with 182+ tools, an AI Code Reviewer integrated into CI/CD, and an Engineering Codex for standards enforcement. A proxy Worker pattern centralizes all LLM traffic, enabling per-user attribution, anonymous tracking, and zero API keys on developer machines. AGENTS.md files generated across 3,900 repos give coding agents structured repo context. The result: 93% R&D adoption, merge request volume nearly doubling quarter-over-quarter, and a security agent processing 7B tokens/day on Workers AI at 77% lower cost than proprietary models.

  13. 13
    Article
    Avatar of agents_digestAgentic Digest·5w

    Meta cuts 8,000 jobs to fund AI infrastructure, Opus 4.7 ships with a hidden token cost

    A developer-focused news roundup covering: Meta's 8,000-person layoff to fund $115–135B AI infrastructure (not financial distress); Claude Opus 4.7's hidden token cost increase (1.46x text, up to 3x images vs 4.6 at same per-token price); Google DeepMind's Aletheia solving 6/10 novel research math problems zero-shot; and a Vercel security breach involving prompt injection and GitHub Actions cache poisoning that put API keys on sale for $2M. Also covers Grok 4.3 beta, MCP hitting 110M monthly downloads, GitHub Copilot running locally via Ollama, Claude Design launch, and growing security debt in AI-generated code.

  14. 14
    Article
    Avatar of arcjetArcjet·4w

    Building a production MCP server in Go

    A detailed walkthrough of how Arcjet built a production MCP server in Go by integrating it directly into their existing API service rather than running it as a standalone sidecar. Key decisions covered include reusing session validation, middleware, and the data layer; designing tools around agent workflows (security briefings, anomaly detection, IP investigation, dry-run impact analysis) rather than simple API wrappers; serving OAuth discovery metadata (RFC 8414 + RFC 9728) while proxying to an external auth provider (WorkOS AuthKit); handling Dynamic Client Registration; writing effective tool descriptions with output shape, workflow hints, and annotations; and separating trusted guidance from untrusted attacker-controlled request data to prevent prompt injection.

  15. 15
    Article
    Avatar of metabaseMetabase·5w

    Metabase AI Hackathon

    Metabase is hosting an AI hackathon open until May 5, 2026, inviting developers to build projects using its new AI tools: an MCP server, Agent API, file-based development, and Metabot for Slack. Participants can work solo or in teams, submit via X or LinkedIn with #MetabaseAI, and two winners will receive a Metabase mechanical keyboard plus a blog feature. All AI features require Metabase version 60 or later.

  16. 16
    Article
    Avatar of lnLaravel News·8w

    PhpStorm 2026.1 Released

    JetBrains has released PhpStorm 2026.1 with a broad set of improvements. Key highlights include Laravel 13, Livewire, and Filament framework support, new Route Search UI, and better Eloquent method handling. PHP 8.5 pipe operator quick-fixes and improved generics type inference are also included. The built-in MCP server now exposes IDE actions to third-party agents like Claude Code and Windsurf, and a Laravel Idea MCP integration is available. AI features include Next Edit Suggestions and the new Junie CLI beta coding agent. Testing improvements cover PHPUnit attribute navigation and Pest test runner enhancements. Developer experience gains include Git worktree support, terminal completion for Git/Docker/kubectl, native Wayland support on Linux, and automatic exclusion of framework-generated directories from indexing. On the frontend side, TypeScript now uses a service-powered type engine by default, with added support for React compiler directives, Vue 3.1.8, Angular 21.x, and modern CSS color functions. Code With Me collaboration service is being sunset after this release.

  17. 17
    Article
    Avatar of supabaseSupabase·8w

    Supabase docs over SSH

    Supabase has launched supabase.sh, a public SSH server that exposes Supabase documentation as a virtual file system of markdown files. AI coding agents can use familiar bash commands like grep, find, ls, and cat to browse and search docs without hallucinating outdated or non-existent APIs. Built on Vercel's just-bash open-source TypeScript Bash emulator, it runs sandboxed without a real shell. The approach complements the existing MCP search_docs tool by enabling exploratory browsing rather than requiring predefined queries. Developers can configure agents via AGENTS.md or CLAUDE.md to check docs before working with Supabase.

See all MCP archives