Bourd, in your agent
Bourd's Model Context Protocol (MCP) server connects your AEO data to Claude, ChatGPT, Cursor, Zed, Copilot, and Gemini CLI. Read citations, run prompts, and draft briefs without leaving your agent.
No API keys. OAuth handles sign-in, refresh, and revocation.
Paste the server URL into your client. OAuth handles the handshake. No API keys, no secrets to store.
Google or email OTP. New to Bourd? Create your account from the agent. 1,000 free credits included.
Your agent now reads from Bourd and writes back. Ask for citations, share of voice, or a content gap scan.
Server URL
Paste this into your MCP client settings.
Remote HTTP server with OAuth 2.1. Plug into any MCP client.
Start with these. Each is a prompt you can paste into your agent today, from first-prompt setup to weekly reporting. Built on eighteen tools spanning prompts, runs, citations, competitors, and insights.
Turn the queries you already rank for into a tagged, intent-clustered prompt set in Bourd.
Generates a brief that converts a citation into a mention, built from the content of every cited page.
Model Context Protocol. An open standard that lets AI assistants call external tools and data sources. Bourd runs an MCP server so agents can read your prompts, citations, competitors, and insights, and write back too: create prompts, apply tags, add competitors, trigger runs.
Claude (web, Desktop, and Code), ChatGPT via connectors, Cursor, Zed, Copilot, and Gemini CLI. Any MCP-compatible client with OAuth support will work.
No. The server uses OAuth 2.1. You sign in with your Bourd account once, and the client handles refresh in the background. No keys to rotate or store.
Yes. Starter, Growth, and Pro all include MCP. No add-on, no Enterprise gate. Credits consumed by MCP calls are billed the same as any other prompt run.
Scoped to your user. One connection covers every account and workspace you belong to. Switch accounts inside a session with the bourd_select_account tool; workspace context is passed per tool call. Your account role limits account-level actions, your workspace role limits workspace-level actions. An agent can never do more than you can.
Reading data (prompts, citations, competitors, insights) is free. Running a prompt response costs the same credits it would cost in the web app: roughly 10 credits for smaller models like Gemini Flash, 100 to 400 for GPT-5 class thinking models.
Everything the agent creates appears in your Bourd dashboard the same way manual work does. Prompt runs, tag changes, competitor updates, and new prompts are attributed to your user, so you can review them alongside anything you created yourself.
Disconnect the Bourd integration in your MCP client. That drops the access and refresh tokens and ends the session.
1,000 free credits on signup. No card. MCP works from the first prompt.