Claude Origin connectorWhat Claude can do in your Origin account.
The Claude Origin connector exposes 2 tools, and both of them read. 0 write tools, 0 left undescribed. You ask about the AI agents running on your company's endpoints, Claude turns the question into a query and reads the answer straight from your Origin data.
Verified Trustpilot reviews · AI, automation & growth agency
Why point Claude at your Origin data
Origin tracks AI agent activity on a company's endpoints: which agents run, who uses them, which prompts go out, which files they touch. With the connector, you stop building a dashboard view for every new question. You ask Claude in plain English, it writes a structured query against your Origin tenant and hands back the rows it found.
Agent activity, prompt usage, model spend. These are the subjects Origin names for natural-language questions. Claude first learns the shape of the data with analytics_describe_schema, then runs analytics_query against your tenant and returns the rows.
Suspicious sessions and compliance audits. Origin lists investigating suspicious sessions and auditing AI usage for compliance among the uses of the connector, on the same data that powers its dashboard.
Shadow AI and adoption reports. The directory sheet also cites hunting unsanctioned AI tools and building adoption reports, from questions such as which agents run and who uses them.
What it won't do: Origin says its server is read-only and cannot change endpoints, agents or alerts, so acting on what you find happens in Origin itself. Nothing runs in the background either, and no alert fires on its own. For other platforms the team has documented, start from the Integrations hub.
Five terms worth knowing
Claude's vocabulary around this connector, minus the jargon.
- Connector
- The link you set up once between Claude and an account you already have, so Claude can use it while it answers.
- Tool
- A named capability the connector opens to Claude. Claude decides when to call it; the directory sheet shows the exact name.
- Authorization
- Origin's own sign-in page, where you give Claude the access it will use later. Granted once per person, and revocable.
- Approval
- The okay Claude asks for in the conversation before it changes anything in an account for you.
- MCP
- The common standard behind connectors: it is what lets an assistant like Claude talk to an outside service.
Set up the Claude Origin connector in three steps
- 01
Open the connector list
In Claude's settings, go to Customize, then Connectors, and look for Origin. On a Team or Enterprise workspace, an Owner or Primary Owner enables it for the organization first, then each member connects individually.
- 02
Sign in with your Origin login
Click Connect on its row. According to Origin, a browser window opens for you to sign in with your Origin account: no API key to create, nothing to paste from the dashboard.
- 03
Approve the authorization screen
Read what Origin's screen lists before you accept. Origin says sign-in goes through your usual Origin login, so there is no bearer token to copy. The access is granted once, for you, and can be taken back the same way.
The 2 tools and what they read
Origin gives Claude 2 tools: 2 that read your account, 0 that change something in it.
Two read tools that work as a pair: one maps your data, the other queries it. Names stay exactly as Claude displays them.
- 2 read
What Claude reads (2)
2 toolsTwo tools that look into your Origin analytics graph without changing anything.
analytics_describe_schema
Returns the full map of your Origin data: entities, fields, metrics, join edges, the query language reference and worked examples. It takes no parameters. Claude calls it to learn how the data is shaped before composing a first query.
analytics_query
Runs a structured query against your company's tenant and brings back the matching rows plus the query's cost. Claude fills in the filters, groupings and aggregations itself from your plain-English request.
What Claude asks before acting
By default, Claude stops and asks for your go-ahead before every action it takes on an account for you. The request appears in the conversation, at the moment it matters.
No source says whether Claude checks with you before calling either Origin tool, so the general rule applies. On Team and Enterprise, workspace owners decide whether a member can let some actions through without a fresh prompt, and they can cap what a connector does for the whole organization. On Origin's side, queries run under your identity: you only see what your Origin role already opens in the dashboard, and Claude sees no more.
Which plans include it
None of the 819 sheets in the official directory publishes availability by plan. For any given connector, that information exists nowhere: it is a genuine gap in the catalog.
The general rule is published: remote connectors are open to all users on Claude, Cowork, Claude Desktop and mobile. On Team and Enterprise, an Owner or Primary Owner turns the connector on for the organization before members connect. For Origin's current status, look at its sheet in the official directory. Good to know: only the publisher and the sheet document this connector; no Claude help article covers it.
Where the connector stops
A connector is not an automation. Claude calls these tools while it answers you: nothing fires when an unknown agent shows up on a laptop.
By default the sheet only exposes reading, and Origin describes its server as unable to modify anything. Read that as the publisher's statement, not a guarantee: an admin can open writes that no public sheet shows. The partner badge is not a security audit, and Anthropic notes on every sheet that it does not choose a publisher's tools. On volume, a query that is too heavy gets rejected, and Origin suggests narrowing it.
Need help connecting Origin to Claude?
A person reads every message.