Connect LurkChain to your assistant
One endpoint works in every assistant below. Pick yours for the exact steps and a snippet to copy. Each guide takes about a minute.
Claude
claude.ai, the desktop app and mobile apps
Claude adds remote MCP servers as custom connectors. Once added, LurkChain is available in chats on every device signed in to your account.
- Open Settings, then Connectors.
- Choose Add custom connector. Name it
LurkChainand paste the endpoint as the remote MCP server URL.https://mcp.lurkchain.com/mcp
- Select Add, then Connect. A browser window opens: sign in to LurkChain and approve read-only access.
- In a chat, open the tools menu and check that LurkChain is switched on. Then ask your question.
ChatGPT
chatgpt.com and the desktop apps
ChatGPT connects to remote MCP servers through Developer mode. You add LurkChain as a connector that signs in with OAuth.
- Open Settings, then Apps and connectors (called Connectors in some versions). Under Advanced settings, switch on Developer mode.
- Choose Create to add a connector. Name it
LurkChain, paste the endpoint as the MCP server URL, and pick OAuth for authentication.https://mcp.lurkchain.com/mcp
- Confirm that you trust the connector, then sign in to LurkChain and approve read-only access.
- In a new chat, open the + menu, turn on Developer mode and select LurkChain. Then ask.
Claude Code
The terminal and IDE extensions
Add LurkChain with one command, then sign in from inside Claude Code.
- Run this in a terminal. Add
--scope userto load LurkChain in every project.claude mcp add --transport http lurkchain https://mcp.lurkchain.com/mcp
- Start Claude Code and type
/mcp. Pick lurkchain, choose Authenticate, and approve in the browser. - Check it is connected:
claude mcp list
Codex
The Codex CLI and IDE extension
Codex reads MCP servers from its config file. The CLI and the IDE extension share it.
- Open
~/.codex/config.tomland add:[mcp_servers.lurkchain] url = "https://mcp.lurkchain.com/mcp"
- Sign in, and approve in the browser:
codex mcp login lurkchain
- Start Codex and type
/mcpto see LurkChain and its tools.
Cursor
Agent chat in the Cursor editor
Cursor reads MCP servers from mcp.json: ~/.cursor/mcp.json for every project, or .cursor/mcp.json inside one project.
- Open the file (create it if it does not exist) and add:
{ "mcpServers": { "lurkchain": { "url": "https://mcp.lurkchain.com/mcp" } } } - Open Cursor Settings and go to the MCP section. LurkChain appears in the list. Select Connect (or Needs login) and sign in.
- Ask a question in Agent chat. Cursor asks before it runs a tool unless you have allowed it.
Gemini CLI
Google's command-line agent
Gemini CLI reads MCP servers from settings.json. Use httpUrl for a streamable HTTP server like LurkChain.
- Open
~/.gemini/settings.json(or.gemini/settings.jsonin a project) and add:{ "mcpServers": { "lurkchain": { "httpUrl": "https://mcp.lurkchain.com/mcp" } } } - Start
geminiand sign in. Approve in the browser./mcp auth lurkchain
- Run
/mcpto see the server and its tools.
Grok
grok.com and the Grok apps
Grok's support for custom connectors is new and still changing, so the menu names you see may differ from these steps.
- Open Grok's settings and find the section for connectors or integrations.
- Add a custom connector and paste the LurkChain endpoint.
https://mcp.lurkchain.com/mcp
- When Grok asks, sign in to LurkChain and approve read-only access.
- Ask a question that needs on-chain data and check that Grok calls LurkChain.
Perplexity
The Perplexity web and desktop apps
Perplexity can use remote MCP servers as connectors. Where the option sits depends on the app and your plan.
- Open Perplexity's settings and go to Connectors.
- Add a custom remote connector and paste the LurkChain endpoint.
https://mcp.lurkchain.com/mcp
- Sign in to LurkChain and approve read-only access.
- Turn the connector on for the search or thread where you want on-chain data.
VS Code
GitHub Copilot Chat in agent mode
VS Code reads MCP servers from .vscode/mcp.json in your workspace. Note the top-level key is servers.
- Create
.vscode/mcp.jsonand add the entry below. Or run MCP: Add Server from the Command Palette and choose HTTP.{ "servers": { "lurkchain": { "type": "http", "url": "https://mcp.lurkchain.com/mcp" } } } - Select Start above the server entry. VS Code asks you to sign in; approve in the browser.
- Open Copilot Chat in Agent mode, open the tools picker and check the LurkChain tools are selected.
Windsurf
Cascade in the Windsurf editor
Windsurf keeps MCP servers in ~/.codeium/windsurf/mcp_config.json. Remote servers use serverUrl.
- Open the file. You can reach it from Windsurf Settings, Cascade, MCP servers, then view raw config. Add:
{ "mcpServers": { "lurkchain": { "serverUrl": "https://mcp.lurkchain.com/mcp" } } } - Save, then refresh the MCP server list in Cascade. Sign in to LurkChain when asked.
- Ask Cascade a question that needs on-chain data.
Any MCP client
Other apps, agents and frameworks
Any client that supports remote MCP servers over streamable HTTP with OAuth can connect. Point it at the endpoint and sign in when asked.
- Add a remote server with this URL:
https://mcp.lurkchain.com/mcp
- If your client only runs local servers, use a bridge such as
mcp-remote, which handles the sign-in for you:{ "mcpServers": { "lurkchain": { "command": "npx", "args": ["-y", "mcp-remote", "https://mcp.lurkchain.com/mcp"] } } } - Writing your own code? Use the REST API with a bearer key instead.
Check it works
Once connected, try one of these. Click to copy. Each answer should end with its cost and your balance.
Stuck on a step?
The help centre covers sign-in, credits and the most common setup problems.