Updated
Connect Cursor to your social media accounts
Yes, Cursor’s agent can post to social media. Add the Postgale MCP server to Cursor with one entry in mcp.json, sign in to Postgale when Cursor asks (or use a Postgale API key), connect the accounts you own in Postgale, and the agent can draft, schedule and publish on Instagram, Facebook, Threads, TikTok, Bluesky and Mastodon and answer comments and messages, while you build.
Can Cursor post to social media?
Postgale is the secure layer for AI agents on social media. MCP (Model Context Protocol) is the open standard that lets AI assistants such as Claude and Cursor use outside tools. An MCP server is what they connect to. Cursor reads MCP servers from a JSON file and supports both MCP sign-in and custom headers, so either way of connecting to Postgale works.
Once connected, the agent can work on Instagram, Facebook Pages, Threads, TikTok, Bluesky and Mastodon, and answer customers on WhatsApp. X, Telegram and Discord are coming soon.
Typical uses: a product team that ships a release and wants the announcement drafted and scheduled from the same session; a developer building their own agent who wants to try the tools before writing code against the REST API.
How do I set up Cursor?
- Create a Postgale workspace and connect the social accounts the agent may use under Dashboard → Accounts.
- In Cursor, open Settings → MCP and add a server, or edit
.cursor/mcp.jsonin the project (or~/.cursor/mcp.jsonfor every project). The URL alone is enough:
{
"mcpServers": {
"postgale": {
"url": "https://postgale.com/mcp"
}
}
}- Cursor shows a sign-in prompt for the server. A Postgale page opens in the browser: sign in, pick the workspace the app may use and approve. The consent page names the app and where it returns to; approve only requests you started yourself.
- Cursor lists the Postgale tools under the server once it connects; enable them for the agent.
- In the agent chat, ask: “Which social accounts are connected in Postgale?” to check the connection.
Can I use an API key instead?
Yes, for shared or automated setups where nobody should sign in personally. In Postgale, open Dashboard → API keys, create a key and copy it (it starts with sk_ and is shown once). Connect the social accounts the agent may use under Dashboard → Accounts. Then add a headers block:
{
"mcpServers": {
"postgale": {
"url": "https://postgale.com/mcp",
"headers": {
"Authorization": "Bearer sk_…"
}
}
}
}Keep the key out of version control: use the global file, or an environment variable if your Cursor version supports ${env:…} in mcp.json.
What should I ask first?
Start with reads, then drafts, then scheduling. The agent sees every connected account by id and picks tools from the list Postgale publishes; you speak in plain language.
- Which social accounts are connected in Postgale?
- Draft an Instagram caption and a shorter Threads post for the fall drop. Save both as drafts, do not publish.
- Schedule the approved drafts for Thursday 9am Eastern on Instagram, Facebook, Threads and our Bluesky account.
- Show me the comments on yesterday’s Instagram post and draft replies to the sizing questions.
- What did last week’s posts do on Instagram and Threads?
A post created without a time or publishNow is saved as a draft and goes nowhere until someone schedules or publishes it. That is the safest default while you learn what the agent does well.
How do I keep a person in the loop?
- Approval: tick “Hold posts from this app for approval” when you sign in, or in the Postgale dashboard require approval for the API key the agent uses (or for the whole workspace, or one account). Posts then wait as pending approval; comment replies and moderation, direct messages, WhatsApp sends and edits to live posts wait in the same queue. A person approves, edits or rejects each one in Dashboard → Approvals; the agent’s own key or sign-in cannot approve.
- Tool annotations: every Postgale tool declares whether it only reads, writes, or publishes live, deletes or spends. Clients that honour annotations ask before destructive tools run.
- Audit log: every write the agent makes is recorded with the key or the connected app, the action and the time.
- Revoking: each sign-in appears under Dashboard → Connected apps and each key under Dashboard → API keys; either can be revoked at any time.
What are the limits?
- The agent only sees accounts their owner connected in Postgale; connecting happens in a browser through the platform’s own sign-in.
- Each platform’s rules still apply and Postgale enforces them: Instagram needs an image or video (JPEG for images), Threads allows 500 characters, Bluesky 300, Mastodon 500 by default, and TikTok needs a privacy level chosen from the creator’s options; direct TikTok posts from an agent always wait for a person to confirm the settings.
- Requests from one key or connected app are limited to 60 per minute by default.
Common questions
Yes. Add the server URL in mcp.json and sign in to Postgale when Cursor asks, or add an Authorization header with a Postgale API key.
Yes, once the accounts are connected in Postgale. Ask for a draft first; publish or schedule when you are happy with it.
Yes. VS Code takes the same server with `"type": "http"`: with only the URL it asks you to sign in, or you can add a headers block with an API key. The exact snippets are in the MCP docs.