Connect AI assistants (MCP)

Claude Code

Add the server once, then check that it connects:

claude mcp add --transport http c1k https://c1k.me/mcp/org_... \
  --header "Authorization: Bearer $C1K_MCP_TOKEN"
claude mcp list

Add --scope user to use it in every project. Then ask, for example: Create a C1K link to https://venture.example/launch with the slug launch26.

Claude (claude.ai and Claude Desktop)

1. The platform owner registers the connector once: Platform, OAuth clients, name Claude, type Public (PKCE only), redirect URLs https://claude.ai/api/mcp/auth_callback and https://claude.com/api/mcp/auth_callback. Its client ID then appears on every organization's Integrations page.
2. In Claude open Settings, Connectors, Add custom connector. Name it C1K and use your endpoint as the URL.
3. Under Advanced settings enter the client ID and leave the secret empty.
4. Choose Connect. Sign in on c1k.me, check the organization and scopes, and allow access.

ChatGPT

1. The platform owner registers the connector: Platform, OAuth clients, name ChatGPT, type Public (PKCE only), redirect URL https://chatgpt.com/connector_platform_oauth_redirect. Its client ID then appears on every organization's Integrations page.
2. In ChatGPT turn on developer mode under Settings, Apps and connectors, Advanced settings, then create a connector with your endpoint as the MCP server URL and OAuth as the authentication.
3. When ChatGPT asks for OAuth client details, enter the client ID and no secret. C1K does not offer dynamic client registration.
4. Sign in on c1k.me and allow access.

Cursor

In ~/.cursor/mcp.json for all projects, or .cursor/mcp.json in one project:

{
  "mcpServers": {
    "c1k": {
      "url": "https://c1k.me/mcp/org_...",
      "headers": { "Authorization": "Bearer ${env:C1K_MCP_TOKEN}" }
    }
  }
}

VS Code (agent mode)

In .vscode/mcp.json. VS Code asks for the token once and stores it securely:

{
  "inputs": [
    { "type": "promptString", "id": "c1k-token", "description": "C1K MCP token", "password": true }
  ],
  "servers": {
    "c1k": {
      "type": "http",
      "url": "https://c1k.me/mcp/org_...",
      "headers": { "Authorization": "Bearer ${input:c1k-token}" }
    }
  }
}

JetBrains IDEs (AI Assistant and Junie)

IntelliJ IDEA, PyCharm, WebStorm, PhpStorm, GoLand, Rider and the other JetBrains IDEs share one setup. For AI Assistant open Settings, Tools, AI Assistant, Model Context Protocol (MCP), add a server and paste the configuration as JSON. For Junie use Settings, Tools, Junie, MCP Settings, or the file .junie/mcp/mcp.json in the project.

{
  "mcpServers": {
    "c1k": {
      "url": "https://c1k.me/mcp/org_...",
      "headers": { "Authorization": "Bearer c1k_mcp_..." }
    }
  }
}

IDE versions that only start local (stdio) servers can reach C1K through the mcp-remote bridge, which needs Node.js:

{
  "mcpServers": {
    "c1k": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://c1k.me/mcp/org_...", "--header", "Authorization:${C1K_AUTH}"],
      "env": { "C1K_AUTH": "Bearer c1k_mcp_..." }
    }
  }
}

The IDE stores this configuration in its settings, so paste the token itself there and treat the settings like a password store.

Codex CLI

In ~/.codex/config.toml; Codex reads the token from the named environment variable:

[mcp_servers.c1k]
url = "https://c1k.me/mcp/org_..."
bearer_token_env_var = "C1K_MCP_TOKEN"

Gemini CLI

Add the server, then check it. Gemini CLI starts MCP servers only in folders you trust, so run it in your project folder and trust the folder when it asks.

gemini mcp add --scope user --transport http c1k https://c1k.me/mcp/org_... \
  --header "Authorization: Bearer $C1K_MCP_TOKEN"
gemini mcp list

The command writes this entry to ~/.gemini/settings.json, which you can also edit by hand:

{
  "mcpServers": {
    "c1k": {
      "httpUrl": "https://c1k.me/mcp/org_...",
      "headers": { "Authorization": "Bearer $C1K_MCP_TOKEN" }
    }
  }
}