About FeedHorizon
FeedHorizon is a hosted social media API and MCP server for AI assistants that manage publishing across connected accounts. It supports drafting, scheduling, queue management, validation, and immediate publishing across several social platforms.
Key features
- 01Manage connected profiles and accounts
- 02Validate posts against platform rules
- 03Draft, schedule, queue, and publish posts
- 04Manage queue slots and post status
- 05Import media and create upload links
- 06Create account connection links
Use cases
- 01Schedule posts across social platforms
- 02Check captions and media before scheduling
- 03Review and refill a publishing queue
MCP support
Provides a hosted MCP server at https://feedhorizon.com/mcp over Streamable HTTP. MCP clients can access tools for post management, platform validation, media handling, and documentation search.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://feedhorizon.com/mcpclaude mcp add --transport http com-feedhorizon https://feedhorizon.com/mcp{"mcpServers":{"com-feedhorizon":{"type":"http","url":"https://feedhorizon.com/mcp"}}}Open in Cursor{"servers":{"com-feedhorizon":{"type":"http","url":"https://feedhorizon.com/mcp"}}}Open in VS Code{"mcpServers":{"com-feedhorizon":{"type":"http","url":"https://feedhorizon.com/mcp"}}}{"mcpServers":{"com-feedhorizon":{"serverUrl":"https://feedhorizon.com/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project READMEFeedHorizon MCP server
Let Cursor, Claude, ChatGPT and other AI assistants schedule and publish social media posts on X (Twitter), LinkedIn, Instagram, Bluesky, Pinterest and TikTok (TikTok in beta: private posts until TikTok's review).
FeedHorizon is a social media API and hosted MCP server. EU hosted (Frankfurt), the first 2 connected accounts are free.
- URL:
https://feedhorizon.com/mcp(streamable HTTP) - Sign-in: OAuth. Your assistant opens a FeedHorizon sign-in page the first time. An API key as
Authorization: Bearer fh_…works too. - Docs: https://feedhorizon.com/docs/guides/mcp
- Official MCP Registry:
com.feedhorizon/mcp
What the assistant can do
- See your profiles and connected accounts
- Check a post against each network's rules (length, media, schedule) before it is saved
- Create drafts, schedule posts at a time, add them to your queue, publish now
- Read and set queue slots, duplicate, cancel and retry posts
- Import media from a URL or create an upload link
- Create a connect link so a client can connect their own account
- Search the FeedHorizon docs
The assistant can do nothing your account could not do through the REST API: same checks, limits and permissions.
Tools
| Tool | What it does |
|---|---|
get_context | Organization, permissions, profiles with time zone and local time, queue state and connected accounts. Called first. |
get_platform_rules | Text length, media count and file types per platform |
search_docs | Searches the FeedHorizon docs |
list_posts | Lists posts, or a calendar range, with filters |
get_post | The full post with per-account status, links and errors |
validate_post | Checks a post against every platform without saving it |
get_queue | A profile's queue slots and which posts hold them |
list_pinterest_boards | The boards of a Pinterest account |
get_tiktok_creator | Creator name and allowed privacy and interaction settings of a TikTok account |
create_post | Saves a draft, schedules a post or adds it to the queue |
update_post | Changes text, accounts, media or time of a post |
publish_post_now | Publishes a post immediately (asks for confirmation) |
cancel_post | Takes a post off the schedule, keeping it as a draft |
retry_post | Publishes the failed accounts of a post again (asks for confirmation) |
duplicate_post | Copies a post as a new draft |
delete_post | Deletes a post in FeedHorizon (asks for confirmation) |
set_queue_slots | Replaces a profile's weekly queue slots |
import_media_from_url | Stores an image, video or PDF from a public URL |
create_upload_link | A link to upload a local file |
create_connect_link | A link to connect a social account to a profile |
Cursor
Add this to .cursor/mcp.json (project) or ~/.cursor/mcp.json (global):
{
"mcpServers": {
"feedhorizon": {
"url": "https://feedhorizon.com/mcp"
}
}
}
Cursor opens the FeedHorizon sign-in when you first use a tool. To use an API key instead, add "headers": { "Authorization": "Bearer fh_your_api_key" }.
Claude, ChatGPT and other clients
Add a custom connector / remote MCP server with the URL https://feedhorizon.com/mcp and sign in. Step-by-step guides for each client: https://feedhorizon.com/docs/guides/mcp
Example prompts
- "Schedule a LinkedIn post for tomorrow 9:00 announcing our new pricing page."
- "Fill next week's queue on X with five tips from our blog post ."
- "Check this Instagram caption and image against Instagram's limits before scheduling."
Links
- Website: https://feedhorizon.com
- Pricing: https://feedhorizon.com/pricing
- API reference: https://feedhorizon.com/docs
- llms.txt: https://feedhorizon.com/llms.txt
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.