Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Rippling MCP ServerWho has pending leave requests that need approval?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Rippling MCP Server
An open-source Model Context Protocol (MCP) server for the Rippling HR/IT/Finance platform. Connect any AI agent or LLM to your Rippling workspace — query employees, manage leave requests, view company structure, and more.
Features
18 tools across 6 domains: Company, Employees, Organization, Leave, Groups, Activity
Bearer token auth — simple API token setup, no OAuth dance required
Rate limit aware — respects Rippling's rate limit headers
AI-friendly errors — structured error messages with actionable fix suggestions
TypeScript — full type safety, built with
@modelcontextprotocol/sdk
Quick Start
1. Get a Rippling API Token
Go to Rippling > Settings > API Tokens > Create API Token. Copy the token.
2. Configure in Claude Desktop
Add to your claude_desktop_config.json:
3. Use It
Ask Claude things like:
"List all employees in Engineering"
"Who has pending leave requests?"
"Show me the company org structure"
"What's Alice's PTO balance?"
Tools Reference
Company
Tool | Description |
| Get company details (name, address, locations) |
| List all departments with hierarchy |
| List work locations and addresses |
Employees
Tool | Description |
| List active employees (paginated) |
| Get a specific employee by ID |
| List all employees including terminated |
| Search by name, email, title, or department |
Organization
Tool | Description |
| List teams and subteam relationships |
| List position levels (IC, Manager, Executive) |
| List custom field definitions |
Leave Management
Tool | Description |
| Get PTO/sick leave balances for an employee |
| List leave requests (filter by status, date, requester) |
| Approve or decline a pending leave request |
| List all leave types configured for the company |
Groups
Tool | Description |
| List all groups |
| Create a new group with members |
| Update group name or members |
| Delete a group |
Activity
Tool | Description |
| Get activity events (hires, changes, terminations) |
Configuration
Variable | Required | Default | Description |
| Yes | — | Your Rippling API token |
| No |
| API base URL |
For sandbox/testing, set RIPPLING_BASE_URL to https://sandbox.rippling.com/api/platform/api.
Development
How It Works
This server wraps the Rippling Platform API (V1) and exposes it through the Model Context Protocol. When an AI agent calls a tool, the server:
Validates the input using Zod schemas
Makes authenticated requests to the Rippling API
Handles rate limiting automatically
Returns structured JSON responses (or actionable error messages)
Limitations
Employee endpoints are read-only (Rippling API V1 restriction)
API token permissions determine which fields are returned per employee
Rate limits are enforced by Rippling (monitor via response headers)
Leave request processing requires appropriate admin permissions
License
MIT
Contributing
Issues and PRs welcome. This is part of the Bifrost MCP infrastructure project.