Plug Claude, GPT, Cursor, Hermes or your own agent into 20 social platforms via an MCP server with 200+ tools, a CLI or a REST API. Schedule, post, run the inbox and analyze content autonomously, no glue code, no per-platform OAuth, no rate-limit babysitting.
Works with Claude · GPT · Cursor · Windsurf · Cline · Continue · Hermes
Pick the integration path that matches your agent architecture. All three share the same workspace, accounts and quota.
Native Model Context Protocol. Connect Claude, Cursor, Windsurf, Cline, Continue and any MCP agent in one config paste.
One npm i -g install for macOS, Linux and Windows. Drop it into any shell, CI job, cron or agent runner with API-key auth.
Standard JSON endpoints at /v1. Use from LangChain, LlamaIndex, AutoGen, CrewAI or any framework.
The MCP server exposes the entire platform, not just posting. From drafts and media to the unified inbox, approvals, per-platform analytics and AI generation, every action in Social media studio is a tool your agent can call. Here are a few of the most-used.
List scheduled and published posts with filtering by status and platform
Retrieve a single post by ID with full details and metadata
Create a new post targeting any connected social media platform
Schedule an existing post for future publishing at a specific time
Remove a post from the workspace by its unique identifier
Generate AI-powered captions optimized for a specific platform and tone
Generate AI content, create a post, and schedule it in a single call
Fetch workspace analytics: post counts by status, platform, and trends
Pull comments, DMs and mentions from the unified inbox across every connected account
Get presigned URLs to upload images and video into the media library, then attach them to posts
Create and update link-in-bio pages, buttons and their click analytics
Approve or reject posts inside the team approval workflow before they publish
Plus 190+ more, across every domain of the product:
Pick MCP, CLI or REST and start posting from your agent.
{
"mcpServers": {
"so-me-studio": {
"url": "https://api.so-me.studio/mcp",
"headers": {
"X-API-Key": "YOUR_API_KEY"
}
}
}
}# install
npm i -g @social-media-scheduler/cli
# auth once
so-me auth:login --api-key $SOME_API_KEY
# schedule a post
so-me posts:create \
--content "Shipping today 🚀" \
--platform LINKEDIN \
--date "2026-07-01T09:00:00Z"await fetch(
'https://api.so-me.studio/v1/posts',
{
method: 'POST',
headers: {
'X-API-Key': 'YOUR_API_KEY',
'Content-Type': 'application/json'
},
body: JSON.stringify({
text: 'Hello from my AI agent!',
socialMedia: 'LINKEDIN',
postType: 'TEXT',
scheduledAt: '2026-07-01T09:00:00Z'
})
}
);MCP-native today. REST-native forever. If your agent can call a tool or hit an endpoint, it can run social media on Social media studio.
From content creation to analytics-driven scheduling, your agent runs the entire workflow on autopilot.
Run AI agents that create, schedule and publish across every social platform on their own, driven by your brand voice and content strategy, no human in the loop required.
Let agents own the entire content calendar, generating, scheduling and optimizing posts for peak engagement windows automatically.
Manage many brands and clients programmatically. Agents switch workspaces and keep distinct brand voices at agency scale.
Agents pull analytics, find top-performing content and auto-schedule similar posts when reach is highest.
From zero to an autonomous posting agent in under five minutes.
Sign up and link any of 20 destinations, Twitter/X, LinkedIn, Instagram, Facebook, TikTok, YouTube, Threads, Pinterest, WhatsApp and more. OAuth is handled once.
Head to Settings → API Keys. The same key works for MCP, CLI and REST.
Paste the MCP config, install the CLI or wire the REST endpoint. Your agent can now create, schedule and analyze posts autonomously.
It is the practice of giving an AI agent (Claude, GPT, Cursor, Hermes or a custom one) programmatic access to your social media accounts so it can draft captions, create posts, schedule them, run the unified inbox and pull analytics without a human in the loop. Social media studio exposes this through an MCP server (200+ tools), a CLI, a REST API and an n8n node across 20 destinations, Twitter/X, LinkedIn, Instagram, TikTok, Facebook, YouTube, Threads, Pinterest, WhatsApp, Dribbble and more.
Anything that speaks MCP (Claude Desktop, Claude Code, Cursor, Windsurf, Cline, Continue, Hermes, custom agents) and any framework that can call a REST API (LangChain, LlamaIndex, AutoGen, CrewAI, OpenAI Assistants, Vertex AI). The CLI works in any shell, CI pipeline or background worker.
Zapier and Make are linear trigger/action workflows for humans. Social media studio is built for agents, model-aware tool definitions over MCP, structured JSON over REST, and an interactive CLI. Your agent decides what to post, when, and why, instead of executing a static recipe.
Yes. Install it with a single `npm i -g @social-media-scheduler/cli`. It authenticates with an API key, prints structured JSON by default and returns proper exit codes, so it drops into GitHub Actions, GitLab CI, cron, Kubernetes Jobs or any runner. Use it for scripted posting, scheduled content campaigns or as a tool inside a larger agent loop.
No. Connect your social accounts once in the dashboard. The MCP server, CLI and REST API all share that connection, your agent never sees a platform token.
Access, prices, and quotas depend on your current plan. See the pricing page for current plans and the API limits table for API call, rate, and webhook quotas.
Yes. Every publish goes through our queue with retries, official-API integrations and per-platform rate-limit awareness. Your agent gets a single create/schedule call and we handle the rest.
Stop juggling tabs, draft, customize, and publish to 20 platforms from a single calendar.