zephex MCP Server
Zephex is a hosted MCP gateway built for AI coding editors. It gives your agent 10 ready-to-use tools — check npm packages for vulnerabilities, audit security headers, read and search code, trace request flows, get project context from any repo, and more. One API key, works instantly with Claude Code, Cursor, VS Code, Windsurf, and others. Free to start at zephex.dev.
Documentation
10 TOOLS · ONE API KEY · ANY AI EDITOR
MCP tools that read your code, search your repo, and audit your packages.
One hosted endpoint. Project context, code search, and package intelligence — instantly available in Cursor, Claude Code, VS Code, and 16 more editors. Free to start.
CONNECT ZEPHEX
TerminalAI Agent
npxpnpmbunnpm -g
>npx zephex setup
get started free
Your AI agent workflows will be 10× faster in minutes with Zephex.
Real catch: check_package flagged a fake Stripe SDK before install — see the workflow
Claude Code
❯ npx zephex setup
✓ Connected · 10 tools live
● check_package — registry scan
↳ Stripe SDK: no typosquat flags
Run setup — 2 min
Works everywhere
One config. Every editor.
27 editors supported. Same tools, same endpoint, zero setup per machine.
get_project_context01/10
01get_project_context02check_package03audit_package04explain_architecture05read_code06find_code07audit_headers08Zephex_dev_info09scope_task10keep_thinking
View all toolsAPI docs
FAQ
Questions beforeyou connect
Quick answers for evaluation, setup, and security review.
01How long does setup take?
Most users connect in a few minutes — create a key, paste one config snippet into your editor, and call the MCP endpoint. No long onboarding, no infrastructure to provision.
02Which clients are supported?
Zephex provides setup guidance for Cursor, Claude Code, VS Code, Windsurf, and JetBrains, and any other MCP-compatible client. If it speaks MCP, it works.
03Do you train AI models on my tool content?
No. Zephex does not use your prompts, tool inputs, tool outputs, or code to train AI models. Your data is routed and enforced at the boundary — nothing is retained for model training.
04How is Zephex different from running an MCP server locally?
A local MCP server is tied to one machine and one editor. You re-install it for every new project, every team member, and every environment. Zephex is hosted — one endpoint that works across all your editors, machines, and teammates from the moment you paste the config.
05How is Zephex different from Context7 or documentation MCP?
Context7 injects fresh third-party library docs into prompts. Zephex reads your actual repo and installed packages — code search, project context, and package audits. Many teams use both: docs MCP for framework APIs, Zephex for shipping in your codebase.
06Is Zephex a memory MCP that saves contacts or chat history?
No. Memory MCPs like Mem0 remember facts across sessions. Zephex is for live engineering: your file tree, dependencies, architecture, and supply-chain checks while you code — not contact saving or long-term chat recall.
07Can Zephex catch a bad npm package before install?
Yes. The check_package tool queries live registry data for typosquats, maintainer signals, and risk flags — the same workflow that flagged a convincing fake Stripe SDK. Follow with audit_package for version-to-version breaking changes.
08How many free MCP requests do I get?
The free tier includes 300 requests per month per API key (one tool call = one request). Pro and Max tiers raise limits — see Pricing for current numbers.
09Where can I read more setup and comparison guides?
See https://zephex.dev/learn/mcp-questions, editor guides under Documentation, and comparisons for vs local MCP, vs Context7, and vs Mem0.
Ready to route your_first endpoint?_
Create API keyRead the docs