Skip to main content

MCP

Connect Lectern to Claude, ChatGPT, Cursor and Codex

Add Lectern’s hosted MCP server to Claude, ChatGPT, Cursor, VS Code or Codex. Sign in with OAuth, then ask what AI answers say about your brand.

Facts and pricing last verified October 21, 2026.

Direct answer

Add https://mcp.trylectern.com/mcp as a remote MCP server in your AI client, sign in to Lectern, and ask what AI answers say about your brand.

What you can ask

Ask in plain language. Your assistant picks the Lectern tool.

  • “How visible is my brand across AI assistants right now?”

    Uses list_sites, check_ai_visibility

  • “Where do AI answers overlook my brand?”

    Uses analyze_visibility_gaps

  • “Which sources do AI answers cite for my category?”

    Uses analyze_citation_sources

  • “Which buyer questions should I prioritize?”

    Uses list_priority_queries

  • “What is my share of the narrative against competitors?”

    Uses calculate_narrative_share, list_competitors

  • “Show my campaigns and how they are performing.”

    Uses list_campaigns, get_campaign_performance

Endpoint and sign-in

MCP endpoint
https://mcp.trylectern.com/mcp

Sign in with your Lectern account through OAuth when your client asks. You choose the team to connect. No API key and no local process are needed. Lectern uses OAuth 2.1 with PKCE and dynamic client registration, so clients register themselves.

Profiles: aeo and full

aeo (default)

A smaller set of 15 research tools: visibility checks, citations, gaps, competitors, crawling, content briefs and campaigns. Good when you mostly want answers.

https://mcp.trylectern.com/mcp?profile=aeo

full

The broader public catalog, including editorial and site tools. It is a bigger menu, not extra permission: your granted scopes, role and approvals still apply.

https://mcp.trylectern.com/mcp?profile=full

You can also send the header X-Lectern-MCP-Profile: full.

Set up your client

Each step list follows the client's own current documentation, linked below it.

Claude Code

  1. Run the command below in your terminal.
  2. Inside a Claude Code session, run /mcp, choose lectern and finish the browser sign-in.
Terminal
claude mcp add --transport http lectern https://mcp.trylectern.com/mcp
Claude Code MCP docs ↗

Claude (claude.ai and Claude Desktop)

  1. Open Customize > Connectors, click + Add, then Add custom connector.
  2. Name it Lectern, paste the endpoint URL and click Continue.
  3. Leave authentication on sign-in (OAuth), click Add, then sign in to Lectern when asked.
  4. In a chat, click + in the lower left, choose Connectors and switch Lectern on.
  5. On Team and Enterprise plans an organization Owner adds the connector first; members then click Connect.
Connector URL
https://mcp.trylectern.com/mcp
Claude custom connectors ↗

ChatGPT

  1. Add a custom MCP app or connector from ChatGPT settings (menu names and plan eligibility change often, so follow OpenAI’s current help article).
  2. Enter the endpoint URL, choose OAuth as the authentication type, and sign in to Lectern when prompted.
  3. Business and Enterprise workspaces may need an admin to allow custom connectors first.
Server URL
https://mcp.trylectern.com/mcp
OpenAI: developer mode and MCP apps ↗

Cursor

  1. Add the entry below to ~/.cursor/mcp.json (all projects) or .cursor/mcp.json (one project).
  2. Reload Cursor, open Customize, and connect lectern. Cursor discovers the OAuth settings from the server and opens the Lectern sign-in.
mcp.json
{
  "mcpServers": {
    "lectern": {
      "url": "https://mcp.trylectern.com/mcp"
    }
  }
}
Cursor MCP docs ↗

VS Code

  1. Add the entry below to .vscode/mcp.json in your workspace, or run MCP: Open User Configuration for all workspaces.
  2. Start the lectern server from the editor prompt above the entry and finish the browser sign-in.
.vscode/mcp.json
{
  "servers": {
    "lectern": {
      "type": "http",
      "url": "https://mcp.trylectern.com/mcp"
    }
  }
}
VS Code MCP configuration reference ↗

Codex

  1. Add the server, then sign in with the second command.
  2. Run codex mcp list (or /mcp in the Codex TUI) to confirm it is connected.
Terminal
codex mcp add lectern --url https://mcp.trylectern.com/mcp
codex mcp login lectern
Codex MCP docs ↗

Check the connection

Ask your assistant:

First prompt
Call Lectern get_connection_info and show my team, sites and available tools.

It is read-only and costs no credits.

What the free plan includes

The Hobby plan is free to start, with $50 in credits and no card. Live visibility checks from your assistant on the free plan are answered with ChatGPT only. If you ask for another engine, the result says which one was used. Plan details are on the pricing page. Some research and generation calls use credits.

Safety and approvals

  • Nothing is published, merged or sent without a person approving it in the Lectern inbox.
  • Deleting something needs a human yes in Lectern, even if your assistant has confirmed it.
  • Granted scopes decide which tools your assistant sees, and every call still checks your role and site access.
  • Lectern reports results with their limits and makes no promise about rankings, traffic or citations.

Troubleshooting

It shows the wrong team
Sign out of the connector in your client and connect again, then choose the right team at sign-in.
It lists no sites
The selected team has no sites yet. Add one in Lectern, or re-authenticate to pick a different team.
A tool is missing
Your scopes or profile may not include it. Switch to the full profile, or reconnect and approve all scopes.
An action says approval is required
Open your Lectern inbox and approve or reject it there.

Try Lectern

See the answers and citations shaping your category.

Start with the free Hobby plan. Lectern turns visibility evidence into a prioritized, approval-gated execution queue.

Start freeSee pricing