Connect a cloud LLM

Connect a cloud app — Claude.ai, ChatGPT Apps, and recent IDE clients — to your runners and their action catalog over remote MCP. emisar signs you in with OAuth, so there's no API key or header to manage.

Claude.ai and ChatGPT Apps ask for a display name and a server URL. emisar is a full OAuth 2.1 authorization server, so the connector discovers it from that URL, registers itself automatically, and sends you through a one-time emisar sign-in. Enter these values:

Emisar

https://emisar.dev/api/mcp/rpc

Claude (claude.ai)#

1. Open Settings → Connectors (on Pro/Max it's Customize → Connectors).
2. Click + → Add custom connector.
3. Enter Emisar as the Connector name, then paste the Remote MCP server URL above into the URL field.
4. Leave Advanced settings (OAuth Client ID / Client Secret) empty — emisar registers your client for you via Dynamic Client Registration. Click Add.
5. Claude opens an emisar tab. Sign in, review what the connector is asking for, and click Authorize. The connector flips on.
6. Optional: open Settings → Connectors → Emisar. Under Tool permissions, set both Read-only tools and Write/delete tools to Always allow. This removes Claude's own per-tool prompts; emisar policy, approvals, and audit still apply.
7. Turn it on for a chat: the + button (lower-left of the composer) → Connectors → toggle emisar.

ChatGPT#

ChatGPT reaches a custom MCP server through Developer mode. Available on all ChatGPT plans.

1. Turn on Developer mode once — it's off by default and flagged Higher risk: Settings → Security and login, toggle Developer mode. (The same toggle sits at the bottom of Settings → Plugins.)
2. Open Settings → Plugins (or chatgpt.com/plugins) and click Create.
3. Fill the form:
   • Name — emisar
   • Description (optional) — e.g. "Run approved infrastructure actions on my fleet."
   • Connection → Server URL, then paste https://emisar.dev/api/mcp/rpc
   • Authentication — choose OAuth
   Check I understand and want to continue, then click Create.
4. ChatGPT sends you to an emisar sign-in → Authorize. On success it lists the tools emisar advertises.
5. Optional — stop the per-call prompts: open emisar under Settings → Plugins, go to Permissions, and switch it from Allow low-risk actions (the default) to Allow all actions. ChatGPT then stops asking before each call; emisar policy, approvals, and audit still apply, so the gate stays on the server, not ChatGPT's prompt.
6. Use it from a new chat: click the + button near the composer, choose More, then select emisar. Seeing tools in settings only proves ChatGPT fetched metadata; this step adds emisar to the conversation context so the model can call it.
7. After changing emisar's tool metadata, open the connector under Settings → Plugins and choose Refresh before testing again.

ChatGPT connects to remote HTTPS MCP servers only — there is no stdio option. The emisar endpoint is already remote, so nothing extra to host.

What the sign-in grants#

The consent screen mints a key bound to your membership, so the connector acts as you — every call is attributed to you and the agent you're using, and it reaches only the runners and actions your own access already allows (how a key inherits its operator's scope). Revoke it any time from LLM agents in the dashboard, or from the connector settings in Claude or ChatGPT.