Origin Is Now Available in the Anthropic Connectors Directory
Today, Origin is officially listed in the Anthropic Connectors Directory, the verified catalog of integrations that bring external tools and data directly into Claude. For teams using Origin to monitor their organization's AI footprint, the telemetry we’re collecting is now accessible right inside your Claude conversations, making it easier for organizations to operationalize their AI observability data.
What it means to be in the Connectors Directory
The Connectors Directory is Anthropic's curated, vetted catalog of integrations built on a Model Context Protocol (MCP), an open standard that allows external tools and data sources to connect directly to Claude and other AI tools. Listings in the directory are reviewed by Anthropic and available across every Claude product, from Claude.ai to Claude Code.
For end users, the experience is conversational. You don't need to know what API to call or which dashboard to navigate to. You simply ask Claude a question about your endpoints and AI usage, and the connected integration provides the context Claude needs to answer it, in real time, from your actual data. Our presence in the directory is validation that Origin has passed Anthropic's review process and is available to any Claude user or organization that wants to connect.
Interacting with Origin in Claude
Once connected, the Origin MCP server gives Claude direct access to the observability data Origin has been collecting across your endpoints. That means the agent inventory, prompt-level activity, usage patterns, and cost data that Origin surfaces in its platform are now queryable through natural language inside the AI tool your team is already using every day.
From understanding your AI investment to how your team is using AI, what previously required deep investigation or was otherwise a black box is now available to query in natural language. This ranges from ranking your departments by token consumption within a given time frame to understanding which prompts fail most frequently.
What it looks like in practice:
Instantly review agent inventory: Ask Claude which AI agents are running across your fleet, which devices they're installed on, and which employees are actively using them.
Understand user behavior in natural language: Origin tracks what agents are running, but as well as what users are prompting them to do and what those agents do in response. Through the connector, you can ask Claude to surface anomalous sessions, characterize a user's topic clusters, or trace the chain of actions from a single prompt to its downstream execution.
Access usage and spend data: Ask why AI projects spiked in cost last week, which teams are driving the most token consumption, or the context around particular projects.
Get started with one device
The fastest way to understand what Origin sees is to experience it on your own machine.
Install Origin on a single device, collect your data, connect the MCP integration from the Anthropic Connectors Directory, and ask Claude what it found. You can find additional support and documentation on how to interface with Origin's MCP server here.
If your organization is deploying AI at scale—be it coding agents, browser agents, productivity tools, or anything in between—you need a clear picture of what's running, who's using it, and what it's doing. This is the operational intelligence Origin provides.
The Origin connector makes it easier than ever to put that visibility to work. Start with one device and connect Origin to Claude today at originhq.com.