Skip to main content

​
What this page covers

Enginy supports any client that can speak remote MCP over Streamable HTTP with OAuth 2.0 — there’s no allowlist of “supported” tools beyond that. This page is a directory: how to connect from each client we document today, and what to expect. If you want the simplest starting point, use Claude (claude.ai) or ChatGPT — most teams, especially non-technical roles like SDRs, connect Enginy from the AI assistant they already chat in day to day. If your team already prefers another client, its setup page has the exact steps. If you need the architecture, start with Overview. Each client below links to its own setup page. If you are debugging auth or transport issues, use Security and troubleshooting.

​
Client directory

​
Practical notes

  • Most successful setups today use remote MCP over HTTP and a browser sign-in flow.
  • Many browser-based flows use a local callback like http://localhost:3118/callback; Claude’s hosted app uses https://claude.ai/api/mcp/auth_callback instead, since it connects from Anthropic’s cloud, not your machine.
  • If a client only supports local stdio connections, it is not a direct fit for Enginy MCP.
  • If you are choosing from scratch, Claude or ChatGPT is usually the fastest way to get started — no terminal, no config file.
  • ChatGPT Developer Mode is not the same thing as the ChatGPT consumer app. The consumer app (Chat and Work modes) only supports OpenAI-verified Apps (Enginy’s verification is in progress), so it is not available there at all. Developer Mode is the separate path that connects custom MCP servers, and it’s in beta across every paid tier — but write access still requires a Business/Enterprise/Edu workspace and an admin opt-in, and the approved tool list freezes at the time an admin approves the connector, so it can go stale as Enginy’s OpenAPI-generated MCP surface evolves.

​
Best fit by workflow

  • Use Claude or ChatGPT when you want your team (especially non-technical roles) working from the AI assistant they already use day to day.
  • Use Claude Code or Codex when your team prefers the CLI — neither is gated by the ChatGPT plan-tier restrictions that apply to Developer Mode.
  • Use Cursor or VS Code when your team already uses project-level mcp.json files.
  • Use Gemini CLI when your team already works in Google’s AI stack and wants MCP in that environment.
  • Use Microsoft 365 Copilot when your org wants one admin-managed connector instead of per-user client setup.
  • Use a bridge only when your preferred client cannot connect to remote MCP directly — see Custom & other MCP clients.

Custom & other MCP clients

Connect a bridge or a homegrown MCP client, plus the technical details behind every client integration.

Security and troubleshooting

Review the auth flow, callback handling, and common failure modes.