Siren MCP
Your agent makes the images, captions, and videos for your brand, then posts them to your socials. You show Siren the brand once. After that you type post this in Cursor, Claude, ChatGPT, Codex, or Grok and it goes out.
Remote MCP server, Streamable HTTP, OAuth 2.1. No API keys to copy.
Connect
Claude Code
claude mcp add --transport http siren https://mcp.mysiren.ai/mcp
Claude (claude.ai): Settings → Connectors → Custom → https://mcp.mysiren.ai/mcp
Cursor: Settings → Tools & MCP → add a remote server, or in ~/.cursor/mcp.json:
{ "mcpServers": { "siren": { "url": "https://mcp.mysiren.ai/mcp" } } }
ChatGPT: Developer Mode on → Settings → Apps & Connectors → Create → https://mcp.mysiren.ai/mcp
Grok: grok.com/connectors → Custom, or grok mcp add --transport http siren https://mcp.mysiren.ai/mcp
The agent opens a consent page on app.mysiren.ai. Approve once. Posting stays off until you turn on Allow posting on that screen. Revoke any time under Developers.
Tools
Posts go through the channels already connected in the Siren dashboard: X, LinkedIn, TikTok and YouTube (Instagram once Meta approves the app). The agent never sees your platform tokens.
Example
> Post an update about the new export feature. Use the dashboard screenshot.
The agent calls clarify_brief, then create_campaign, waits on get_run, and calls post_run. One image or film, a caption, and the post, on brand every time.
This repo
Listing and connect docs. The server is hosted at mcp.mysiren.ai. Questions: open an issue or write to [email protected].
Siren · a thousand voices, one signature.