rotacloud-mcp-node
MCP server exposing the RotaCloud API to Claude and other MCP clients, covering all 129 documented v1 operations across 36 resources including shifts, attendance, leave, users, locations, roles, and timesheets.
README
rotacloud-mcp-node
An MCP server exposing the RotaCloud API to Claude and other MCP clients. It covers all 129 documented v1 operations across 36 resources — shifts, attendance, leave, users, locations, roles, timesheets and the rest.
Tools are generated from RotaCloud's published OpenAPI specification (vendor/openapi.json), so parameter names, types and descriptions come straight from the documentation.
Installation
As a Claude Desktop extension
Download rotacloud-mcp-node.mcpb and open it with Claude Desktop. You will be prompted for your RotaCloud API key, which you can generate from within your RotaCloud account.
Manual / development
npm install
export ROTACLOUD_API_KEY="your-api-key-here"
node server/index.js
Add it to claude_desktop_config.json:
{
"mcpServers": {
"rotacloud": {
"command": "node",
"args": ["/absolute/path/to/rotacloud-mcp-node/server/index.js"],
"env": {
"ROTACLOUD_API_KEY": "your-api-key-here"
}
}
}
}
Configuration
| Variable | Required | Purpose |
|---|---|---|
ROTACLOUD_API_KEY |
yes | API key generated from your RotaCloud account |
ROTACLOUD_USER_ID |
no | Act on behalf of this user by default (sends the User header) |
By default requests are made as an anonymous user with administrative permissions. Setting ROTACLOUD_USER_ID makes every request act as that user instead. Tools whose behaviour depends on the acting user — me_*, messages_*, leave_requests_*, swap_requests_* and unavailability_requests_* — also accept an as_user argument to override it per call.
Tools
Tools are named {resource}_{action}, e.g. shifts_list, shifts_create, users_retrieve, leave_requests_approve.
Names are derived from each operation's summary in the API documentation rather than from its HTTP method, because in this API the two often disagree — DELETE /users_clocked_in/{id} clocks a user out, and POST /swap_requests/{id} denies a swap. The tool names reflect what the operation actually does: users_clocked_in_clock_out, swap_requests_deny_shift_admin.
Dates and times
RotaCloud mixes three formats, and the tools follow the API exactly:
- Unix epoch seconds for shift and attendance times (
start_time,in_time, and thestart/endrange filters on/shifts,/attendance,/availability,/pay_periods…). These tools also accept an ISO 8601 string and convert it for you. YYYY-MM-DDstrings for leave, day notes, TOIL and user dates (start_date,end_date,dob…).HH:MMstrings for logbook event times and availability windows.
Pagination
List endpoints accept limit and offset. Responses that are paginated are returned as:
{
"meta": { "total_count": 137, "links": { "next": "…", "last": "…" } },
"data": [ … ]
}
Results are not auto-paginated — each call returns one page, so a wide date range cannot flood the context. Follow meta.links.next or increment offset to page through.
Request bodies
Write tools list every field documented for that endpoint, but accept unknown fields too. RotaCloud's published body schemas are derived from example payloads and under-describe reality (for instance role_rates is documented with the example's literal role IDs as keys), so rejecting undocumented fields would block valid writes. Each write tool's description includes the documented example payload.
Resources
accounts(1)attendance(5)attendance_approved(2)availability(2)day_notes(5)days_off(3)days_off_patterns(5)documents(6)groups(5)holiday_allowances(2)holiday_allowances_custom(3)leave(5)leave_embargoes(5)leave_requests(6)leave_types(1)locations(5)logbook_categories(5)logbook_events(5)me(2)messages(2)pay_periods(3)pins(1)roles(5)settings(1)shifts(5)shifts_acknowledged(1)shifts_published(2)swap_requests(5)terminals(5)terminals_active(3)timezones(2)toil_accruals(4)toil_allowance(1)unavailability_requests(6)users(5)users_clocked_in(5)
Scope
This server covers the v1 API as published at https://rotacloud-api-docs.netlify.app/. RotaCloud's official Node SDK exposes some additional v2 endpoints (invoices, v2 logbook, user onboarding) which are not part of the public documentation and are not included here.
Regenerating
server/tools.js is generated and committed. To pick up a newer version of the API:
curl -o vendor/openapi.json https://rotacloud-api-docs.netlify.app/openapi.json
npm run generate
The generator fails loudly if two operations would produce the same tool name.
Building
npm run build # mcpb pack
Check the contents of the resulting .mcpb before distributing it — mcpb pack sweeps local dotfiles into the bundle.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.