Step 1
Open your AI client's connector settings
In Codex, Claude, or another MCP-compatible client, open the place where remote MCP servers or connectors are managed.
AI agent setup
Use the official DoroSocial MCP server to let an AI agent inspect approved brand context, prepare social content, manage drafts, and check publishing workflows through a secure DoroSocial connection.
Before you start
You need a DoroSocial account and an MCP-compatible AI client that supports remote Streamable HTTP servers.
Access model
DoroSocial asks you to sign in, choose brands, and approve the agent connection.
Best first prompt
Ask the agent to inspect brand context and suggest a content plan before creating drafts.
Step 1
In Codex, Claude, or another MCP-compatible client, open the place where remote MCP servers or connectors are managed.
Step 2
Select Streamable HTTP or the equivalent remote MCP server type. Studio is for local/custom development workflows, not the normal DoroSocial hosted connection.
Step 3
Name the connector DoroSocial, use https://doro.social/mcp, and leave bearer token, custom headers, and headers from environment variables empty for the standard OAuth flow.
Step 4
Click Authenticate. Sign in to DoroSocial if prompted, then review the agent connection request.
Step 5
Select only the brands the agent should work with, review the requested DoroSocial capabilities, and approve the connection.
Step 6
Start with a clear task such as: Review my DoroSocial content calendar and draft next week's posts for the selected brand.
Open DoroSocial Settings, go to Agent connections, and revoke the connection you no longer want to use. You can reconnect later and choose a different brand set.
The AI agent workflow page explains the user-facing value: connect an AI agent to DoroSocial so it can help plan, draft, review, and manage social content.