Thanks for agreeing to test the upcoming Streamtime Public MCP Server. We’ve been working on it for a few months now, and we’re really proud of how far it’s come. So of course, we’d like to hear what you think.
You’re in an early notification testing group for testing our public MCP server. You will be connecting your Streamtime to an AI assistant via MCP and any changes you make in your connected LLM will be reflected in your live account. Be careful with any ‘destructive’ behaviour: as with all Streamtime data, we can’t roll back any unexpected deletions or mistakes.
You’ll be seeing how the tool interacts with your questions and prompts so that we can refine the stability and ensure we’re building the right tools for a real workflow.
We’d love for you to document both hits and misses: what did you do that worked for you, and aided your workflow? And what did you try that didn’t succeed? We’ll hit you up for consolidated feedback in a week or two to hear how it’s going, but don’t be shy if you want to proactively reach out.
You can find more information in our public help centre here.
Below are the details to connect to Streamtime’s MCP Server via Claude Desktop. If you’re using a different tool, the same connector information applies.
The MCP Server URL is: **https://mcp.streamtime.net/**
https://www.loom.com/share/42e170f1c991409bb4e6ba135f6badc1
For instructions to connect to other tools, check out this article: Connecting Streamtime to Your AI Assistant (via MCP)
For a Quick Start Guide to what the MCP can and can’t do, check out our article: The Streamtime MCP: Quick Start Guide
The Streamtime Connector acts as a login for each team member, and has access to data that that team member can see. For example, if the team member connecting hasn’t got Cost Rate access, that team member won’t be able to access data that they can’t see in Streamtime.
The connector currently has access to a subset of tools to support most common workflows requested by users in the early research phase.