MCP connectorsRippling

Rippling MCP: connect Rippling to Claude and ChatGPT

Ask Claude or ChatGPT how many people each department has, who reports to a manager or which new hires start next month. The answer comes straight from Rippling, without building a report first.

Rippling has its own MCP server, and for many teams it is the place to start. We build a connector on Rippling’s API when Rippling is one of several systems, or when you want company-specific tools. It only reads, your company decides which groups get it, and pay stays out unless you choose otherwise.

Built for your company · Works in Claude, ChatGPT and Copilot

In practice

What your team can do with Rippling in Claude and ChatGPT

HR

Headcount in plain words

Workers by department, entity, title or work location, with start dates and managers, asked instead of built as a report.
Managers

Who is on my team

Your reports, their titles and start dates, and who joins next month, in one answer before the team meeting.
Operations and finance

Headcount next to the plan

People per entity and department side by side with your finance and planning systems, in the same chat.
IT

Joiners and leavers

Who starts and who leaves in the coming weeks, so accounts and equipment are ready in time.
Examples

Ask it like this

  • “How many people do we have in each department, by work location?”

  • “Who reports to Jason Miller, and when did each of them start?”

  • “Which new hires start in December, and who is their manager?”

  • “Who has left the company since September?”

  • “List everyone in the Austin office with the title Account Executive.”

  • “Which teams does Ashley Rivera belong to?”

The connector

What we build for Rippling

Deployed OS has no ready Rippling connector yet. We build it for your company when you start, on Rippling’s own API, with the same rules as every other connector: each person’s own login, access decided by group, every call logged.

Workers
People with the fields your company allows, such as title, department, manager, work location, start date and status.
Organisation
Departments, teams, titles, legal entities and work locations, so the AI can group people the way your company does.
Custom fields and objects
The fields and objects your company has added in Rippling, read the way you have set them up.
Time
Recorded working time from Rippling’s Time API, per person and period, if you use Rippling Time.
Security

Who sees what

Read-only
The connector reads. It does not hire, change a worker, approve anything or touch payroll.
A company token, access per group
Rippling’s API connects with a token created in Rippling’s Developer app, with the scopes you choose, so people do not sign in with their own Rippling login. Deployed OS decides who may use it: admins choose which groups get Rippling, and a group without access never sees the connector.
Pay kept apart
The token only reaches the scopes chosen when it is created. We leave compensation and other sensitive fields out unless you ask for them, and then only for the groups you name, such as HR or payroll.
Every call logged
Who asked, what was read and when. Deployed OS keeps the log, and it can be exported.
Setup

Set up the Rippling connector in three steps

01
We go through with you which questions matter most and who should be able to ask them, and decide together which Rippling data the connector may read.
02
Your Rippling admin creates an API token in the Developer app with read scopes for those areas, and you choose which groups get Rippling in Deployed OS.
03
People in those groups add your company’s Deployed OS connection in Claude or ChatGPT. From then on they just ask.
Compared

Rippling’s own MCP server or Deployed OS?

Rippling has its own MCP server, hosted by Rippling and on by default for every company. Each person connects with their own Rippling sign-in and cannot reach more than they can in Rippling, and admins set which tools are allowed in Rippling’s MCP Gateway. It works with Claude, ChatGPT and other MCP clients, and using it needs an active Rippling AI subscription. If you have that subscription and want Rippling in your AI tool, it is a good fit.

Deployed OS is for when Rippling is one system among several, or you want tools built for your company, such as a joiners and leavers view for IT that combines Rippling with your other systems. It sits next to your other tools in one company connection, admins decide which groups get each system, every call is logged in one place, and the same rules apply in Claude and ChatGPT, next to your company’s skills and guardrails.

Their documentation: Rippling MCP

Questions

Rippling MCP: common questions

Yes. Rippling has its own MCP server, managed in Rippling’s MCP Gateway, where each person works within their own Rippling permissions. It needs a Rippling AI subscription. Deployed OS adds Rippling to one company connection with your other systems, with access per group and one log.

Not through the connector we build. It reads workers, the organisation, custom data and time. It does not hire, edit records, approve requests or touch payroll.

Nobody, unless you decide otherwise. We leave compensation out of the token’s scopes. If pay should be readable, we keep it to the groups you name, such as HR or payroll, and every call is logged.

Not with the connector we build. Rippling’s API token belongs to the company setup and reaches only what its scopes and its creator’s permissions allow. Access is set in Deployed OS instead. If each person’s own Rippling permissions matter most, Rippling’s own MCP server does that.

Deployed is an EU company and fully GDPR compliant, and you choose where your data is stored, down to the city. Rippling data is fetched when someone asks for it.

It depends on the size of your team. Get in touch and we will walk you through it.

More connectorsAll connectors
Let’s talk

See it on your own tools.

Tell us which systems your company runs on. We’ll show you what your team can ask, and what IT gets to control.