Raycast browser MCP setup

Connect E2LLM to Raycast

Give Raycast access to the browser session you already use. E2LLM connects as a remote MCP server, so Raycast can read the rendered page, inspect controls, and act through explicit browser tools.

Remote MCPStreamable HTTPDynamic OAuthRaycast Pro

Ready-made setup instruction

Paste this into Raycast AI or follow it directly in Root Search.

Configure E2LLM as a remote HTTP MCP server in Raycast.

1. Open the Install MCP Server command.
2. Name the server "e2llm", choose HTTP, and set the URL to https://mcp.e2llm.com/mcp.
3. Choose Dynamic OAuth and leave HTTP Headers empty.
4. Install the server, select Sign In, and complete OAuth.
5. Wait for the tool list to load, then ask me to run the verification prompt from this page with @e2llm.

Raycast exposes each connected server through an @-mention and an Ask command.

Connect

Connect your browser

Open the E2LLM setup wizard, install the browser extension, and sign in.

Install the HTTP server

In Raycast Root Search, run Install MCP Server. Use these fields:

Name: e2llm
Transport: HTTP
URL: https://mcp.e2llm.com/mcp
OAuth Type: Dynamic

Sign in and load tools

Install the server, choose Sign In, and complete OAuth. Raycast stores the token encrypted and loads the tool list.

Official client mechanics: Raycast lists MCP as a Pro feature and supports Streamable HTTP, Dynamic OAuth with PKCE, encrypted tokens, server @-mentions, and MCP tools in AI Chat and Quick AI. Read Raycast's MCP documentation →

Verify the connection

Confirm that Raycast lists the E2LLM tools. Open a normal webpage in the connected browser and run this read-only check:

Use E2LLM to list my open browser tabs. Then capture the active tab and tell me its page title and primary heading. Keep the page unchanged.

Pass condition: Raycast returns the tabs, captures the active page, and reports details from that page through real tool calls.

Worked use case

Build a sourced comparison from several open product tabs

Raycast is available from anywhere on the desktop. E2LLM lets a scoped AI Chat read the product pages already open in your browser.

Use @e2llm to compare the product pages currently open in my browser.

1. List the tabs and identify the relevant product pages.
2. Read each page's product name, stated price, primary feature, and documented limitation.
3. Keep every claim tied to the tab title and URL where it appeared.
4. Return a compact comparison table and a short list of unresolved questions.
5. Keep every browser tab unchanged.

This workflow turns open research tabs into a structured, source-linked comparison from Raycast AI.

Troubleshoot the connection

The server is absent: run Manage MCP Servers and confirm e2llm is installed.
Tools are unavailable: select the server, choose Start or Restart, and wait for the tool list.
OAuth needs another attempt: use Logout Server, then Sign In again with Dynamic OAuth.
No tabs appear: connect the browser extension with the same account.
The agent chooses other tools: type @e2llm to scope the request to this server.

Capabilities

E2LLM exposes the tools advertised by the running production deployment: structured page capture, targeted reads, tab discovery, exploration, and explicit browser actions. The generated MCP tool reference is built from that named production deployment and stays aligned with what it advertises.