---
source: https://autoadify.com/mcp/chatgpt
title: "Autoadify in ChatGPT"
---

# Schedule social media posts from ChatGPT

> Add https://mcp.autoadify.com/mcp to ChatGPT and sign in to Autoadify. ChatGPT can then write posts, generate images and videos, and schedule them to Instagram, Facebook, LinkedIn, X, TikTok, YouTube, Threads, Bluesky, Pinterest and Reddit. No API key. Included on Autoadify plans from Pro.

Ask ChatGPT to turn a blog link into posts for every platform, generate an image for each, and schedule them in Autoadify.

## How to connect Autoadify to ChatGPT

1. In ChatGPT settings, open Apps & Connectors → Advanced and turn on Developer mode.
2. Add a custom MCP server: name it Autoadify, paste https://mcp.autoadify.com/mcp, and choose OAuth.
3. Sign in to Autoadify and approve the permissions.
4. In a chat, click + → Developer mode and switch Autoadify on.

**Where it works:** ChatGPT Plus, Pro, Business and Enterprise, with Developer mode turned on.

- On ChatGPT Business and Enterprise, a workspace admin may need to allow custom connectors first.
- ChatGPT asks before it runs a tool that changes something, such as scheduling a post.

## What ChatGPT can do once connected

- `whoami` (read) - The workspace, plan, permissions and remaining quota of the key.
- `list_channels` (read) - Connected channels with each platform's posting rules — media required, image limits, text length, required fields.
- `create_post` (publish) - Schedules a post to one or many channels, publishes it now, or saves it as a draft for you to review. Per-platform text, media, Reddit and TikTok options.
- `list_posts` (read) - Posts with their status on each channel, filtered by status, channel or date.
- `get_post` (read) - One post: status per channel, the live URL, and the error if a channel failed.
- `update_post` (publish) - Edits a scheduled post or draft: text, per-platform text or time.
- `delete_post` (publish) - Cancels a scheduled post.
- `list_media` (read) - Searches the media library by name, alt text and a description of what each file shows.
- `get_media` (read) - One file: status, URL, type, size and dimensions.
- `import_media` (publish) - Pulls an image, video or PDF from a public URL into the library.
- `create_media_upload` (publish) - A one-time URL for uploading a local file, for agents with a shell.
- `complete_media_upload` (publish) - Finishes an upload once the file has been sent.
- `delete_media` (publish) - Deletes a file; refused while a scheduled post still uses it.
- `list_ai_models` (generate) - The image and video models, their credit cost, which ones your plan includes, and your credit balance.
- `generate_image` (generate) - Makes an AI image and saves it to the media library, ready to attach to a post.
- `generate_video` (generate) - Starts an AI video; it lands in the media library in 1–10 minutes depending on the model.
- `get_generation` (read) - Whether an image or video is done, with the media id to attach. Failed ones are refunded.

You choose Read, Generate and Publish when you approve. Ask for drafts to review posts in Autoadify before they go out. Disconnect any time from the AI assistants page in Autoadify.

Other assistants: [Claude](https://autoadify.com/mcp/claude), [Grok](https://autoadify.com/mcp/grok), [Perplexity](https://autoadify.com/mcp/perplexity), [Manus](https://autoadify.com/mcp/manus), [Claude Code](https://autoadify.com/mcp/claude-code), [Codex](https://autoadify.com/mcp/codex), [Cursor](https://autoadify.com/mcp/cursor), [Gemini CLI](https://autoadify.com/mcp/gemini), [OpenClaw](https://autoadify.com/mcp/openclaw), [Hermes Agent](https://autoadify.com/mcp/hermes).

---

_Source: https://autoadify.com/mcp/chatgpt_