Summary
This is a clean architecture pattern for building MCP servers in Next.js that keeps your route handlers thin and your logic reusable. The core idea is smart: define your tools once with Zod schemas in lib files, then share that exact code between your MCP endpoints and Next.js server actions. That means your web UI for testing and your actual MCP tools stay in sync automatically. It uses mcp-handler for HTTP transport, works with mcp-remote for Claude Desktop, and the folder structure is straightforward App Router stuff. If you're shipping MCP tools on Vercel or want a proper web interface alongside your AI integrations, this gives you a solid starting point without the usual duplication headaches.
Install to Claude Code
npx -y skills add gocallum/nextjs16-agent-skills --skill mcp-server-skills --agent claude-codeInstalls into .claude/skills of the current project.
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →Files
SKILL.md
Select a file.
Featured
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →
