mcpField
S
Smart Post Studio

Smart Post Studio

Remote
smartpoststudio.comSocial Media0Updated 2w ago

About Smart Post Studio

Smart Post Studio is a hosted social media scheduler for people managing posts across eight platforms. Its remote MCP server lets compatible assistants read queues, create drafts, and schedule posts, with no post-now tool.

Key features

  1. 01Read and filter the post queue
  2. 02Create drafts or scheduled posts
  3. 03Edit or cancel scheduled posts
  4. 04Check per-platform publishing results
  5. 05Manage permissions by connection

Use cases

  1. 01Review posts scheduled across platforms
  2. 02Draft platform-specific versions of a post
  3. 03Schedule a draft or move its publishing time
  4. 04Check whether a post published successfully

MCP support

The hosted remote MCP server uses Streamable HTTP and supports OAuth or Bearer token authentication. It exposes tools for queue access and post creation, editing, scheduling, cancellation, and results; it has no immediate-publishing tool.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://app.smartpoststudio.com/v1/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

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 README

Smart Post Studio MCP — connect your AI to your social media queue

Smart Post Studio is a social media scheduler for 8 platforms (Instagram, Threads, TikTok, X, Facebook, YouTube, LinkedIn incl. Company Pages, Fanvue). It exposes a remote MCP server so any MCP-capable assistant — Claude, Claude Code, Cursor, VS Code + Copilot, Gemini CLI — can read your queue, draft posts and schedule them.

This repo is the public documentation and example configs for that server. The server itself is hosted; there is nothing to install or run.

Server URL:  https://app.smartpoststudio.com/v1/mcp
Transport:   Streamable HTTP
Auth:        OAuth (browser sign-in) or Bearer token
Plan:        included on every paid plan, from $8/mo · 14-day free trial, no card

Full guide: https://smartpoststudio.com/mcp


60-second setup

Claude (claude.ai or Claude Desktop)

Settings → Connectors → Add custom connector

  • Name: Smart Post Studio
  • URL: https://app.smartpoststudio.com/v1/mcp
  • Leave OAuth Client ID / Secret empty

Click Add → sign in with your Smart Post Studio email/password → choose what the AI may do (read, draft, schedule) → Allow.

Claude Code
claude mcp add sps --transport http https://app.smartpoststudio.com/v1/mcp

Then run /mcp and finish the sign-in in your browser.

Cursor / VS Code / any client that uses headers

Create a token in the web app → AI access → Create token (shown once), then see examples/ for a ready-made config:

{ "mcpServers": { "sps": {
  "url": "https://app.smartpoststudio.com/v1/mcp",
  "headers": { "Authorization": "Bearer YOUR_TOKEN" } } } }
Test it

Ask: "What's in my Smart Post Studio queue?"


Tools

ToolWhat it doesNotes
get_connected_platformsLists the platforms you've connected and whether each publishes from the cloud or from your desktop appThe AI can never connect a platform itself
list_postsReads your queue, filterable by status (e.g. scheduled)Newest first
create_postCreates a draft or scheduled post: caption, platforms[], optional media_urls[], optional scheduled_atPer-platform limits enforced server-side (X ≤ 280 chars, Threads ≤ 500, Instagram requires media)
update_postEdits caption, platforms, media or timeRequires expected_updated_at — optimistic locking, so two agents can't clobber each other
cancel_postReturns a scheduled post to draftsNot a delete
get_post_resultsPer-platform publish result for a postUse it for "did yesterday's post go out?"

Platform keys: instagram, threads, tiktok, x, facebook, youtube, linkedin (personal), linkedin_page (Company Page), fanvue.

Guardrails (by design)

  • There is no "post now" tool. Everything an AI creates is a draft or a scheduled post at least 10 minutes out, visible in your Queue with an AI badge. That is a window to intervene, not an approval step.
  • Permissions are per connection: read / draft / schedule. Start with read + draft.
  • Rate limits: 30 requests/min, 100 AI-created posts/day.
  • Media by URL only, public https, ≤ 25 MB each.
  • Tokens last 90 days, max 5 per account, revocable instantly under AI access.
  • A single post can't mix cloud-published and desktop-published platforms (mixed_fire_platforms) — split it in two.
  • Every action is written to your account history.

Error reference

ErrorMeaning
mcp_not_enabledNo active paid plan on the account
platform_not_connectedConnect it in the web or desktop app first
scheduled_at_too_soonPick a time ≥ 10 minutes ahead
caption_too_longOver the platform limit (X 280 raw characters, Threads 500)
mixed_fire_platformsSplit into one cloud post and one desktop post
rate_limitedWait a minute and retry
media_url_too_largeImage/video over 25 MB
unauthorizedToken revoked or expired — create a new one

Example prompts

See examples/prompts.md for a full week. The short version:

  • "What's scheduled across all my platforms for the next seven days? Group it by day and flag any empty day."
  • "Draft posts about [topic] for Instagram, Threads and LinkedIn — same idea, different openings. Keep Threads under 500 characters. Save as drafts, don't schedule."
  • "Schedule the Instagram draft for Tuesday 10:00 with this image: https://…"
  • "Move that post to Friday." / "Cancel it."
  • "Did yesterday's post go out everywhere?"

Which clients can write?

Tested: Claude (all tiers incl. free), Claude Code, Cursor — read + write. ChatGPT: full MCP including writes is Business/Enterprise/Edu only per OpenAI's docs; Pro is read-only. Details and dates: Which AI can actually schedule your posts over MCP?

Links

License

Documentation and examples in this repository: MIT. The Smart Post Studio service is proprietary.

Comments

Markdown supported

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.