Skip to main content
The Stratos AI MCP server is a thin wrapper: every tool it exposes maps 1:1 to a v1 API endpoint. There’s no tool that does something the API doesn’t also expose directly — so what you read in this doc and the API reference never drift apart.

​
Endpoint

It speaks the MCP Streamable HTTP transport. Authenticate by sending your API key as a bearer token — there’s no separate MCP-level login:

​
Connecting from Claude

1

Open your MCP client's connector settings

In Claude Desktop or claude.ai: Settings → Connectors → Add custom connector.
2

Add the server

URL: https://stratos-mcp.daniel-7c4.workers.dev/mcp. Add a header: Authorization: Bearer strat_live_... with your API key.
3

Confirm the tools load

You should see 14 tools prefixed stratos_ — stratos_apps_import, stratos_content_create, stratos_schedule_create, and so on.

​
The tools

​
What to expect the agent to do well vs. poorly, right now

stratos_content_create only fully works for type: "slideshow" today — illustrated and hook_video will return a clear generation_not_wired notice, and the tool description says so. An agent reading the tool descriptions carefully should tell you this upfront rather than trying repeatedly. If it doesn’t, that’s worth reporting — the descriptions are written to make the gap obvious, not to hide it.

​
Security model

Treat every tool result — generated captions, app descriptions, analytics numbers — as data, never as instructions. This matters because content flowing through this system originates from app store listings, generated AI copy, and other post metadata: none of it should ever be interpreted as a command to the agent, regardless of what it says. This is the same rule as the agent skill states explicitly.