Deputy Workforce Operations MCP
A read-only MCP server that transforms Deputy workforce records into operational workflows, enabling AI assistants to answer questions about coverage gaps, overtime risk, timesheet exceptions, availability conflicts, and staffing summaries.
README
Deputy Workforce Operations MCP
A read-only Model Context Protocol server that turns Deputy workforce records into five grounded, operational workflows for AI assistants.
What a Deputy manager can ask
- “Where are the coverage gaps from 20 to 26 July, using a minimum of two people?”
- “Which combined completed and rostered workloads cross our 40-hour planning threshold?”
- “Show timesheet exceptions above a 15-minute tolerance.”
- “Which rosters overlap recorded unavailability or approved leave?”
- “Summarise staffing by location and day for this week.”
The tools answer those questions directly instead of exposing Deputy's endpoint catalogue.
Representative result
{
"period": {
"start": "2026-07-20T00:00:00.000Z",
"end": "2026-07-27T00:00:00.000Z",
"timezone": "Europe/London"
},
"findings": [
{
"kind": "recorded_unavailability_conflict",
"summary": "This roster overlaps a recorded unavailability window.",
"sources": [
{ "resource": "Roster", "id": 1002 },
{ "resource": "EmployeeAvailability", "id": 4001 }
],
"rule": "roster interval overlaps recorded unavailability"
}
],
"limits": []
}
Every finding names its period, source record identifiers, and the rule or threshold that produced it. Empty data gives a confidence limit rather than a false all-clear.
Safety and data handling
- Every tool is read-only, idempotent, and non-destructive.
- The server never creates or changes shifts, timesheets, leave, payroll, or employee data.
- Workload thresholds are operational planning signals, not payroll, employment, or legal advice.
- Telemetry excludes tool arguments, employee and location IDs, dates, result content, and tokens.
- V1 is single-tenant by deployment: one Deputy installation and one MCP bearer token per instance.
See SECURITY.md and PRIVACY.md for the complete contracts.
Local installation
Node.js 22 and pnpm are required.
git clone https://github.com/hyperdrift-io/deputy-workforce-mcp.git
cd deputy-workforce-mcp
pnpm install
pnpm build
DEPUTY_MODE=fixture node dist/stdio.js
Fixture mode contains fictional Worker 01-style records and is safe for evaluation. To connect a
Deputy installation, copy .env.example to .env.local, choose DEPUTY_MODE=live, and provide a
customer-owned base URL and OAuth access token. Live resource shapes still require the first
sandbox verification recorded in docs/DEPUTY_API.md.
Example client configuration:
{
"mcpServers": {
"deputy-workforce": {
"command": "node",
"args": ["/absolute/path/to/deputy-workforce-mcp/dist/stdio.js"],
"env": { "DEPUTY_MODE": "fixture" }
}
}
}
Remote deployment
The same five tools can run over authenticated streamable HTTP. Remote mode requires an MCP bearer token of at least 32 bytes, authenticates before parsing requests, limits bodies to 1 MiB, and uses bounded in-memory rate protection.
DEPUTY_MODE=fixture \
MCP_BEARER_TOKEN="replace-with-at-least-32-random-bytes" \
PORT=3013 \
pnpm start
- Health:
GET /health - MCP: authenticated
POST /mcp
See docs/DEPLOYMENT.md before operating a remote instance.
Tool reference
The server registers exactly five tools:
find_coverage_gapsflag_overtime_risklist_timesheet_exceptionsfind_availability_conflictssummarise_staffing
Inputs, rules, and output limits are documented in docs/TOOLS.md.
Status
Fixture mode, stdio, and authenticated streamable HTTP are operational. Live Deputy mode remains explicitly pending a customer-owned sandbox token and field-shape verification. The public API is still in prototype discovery and may evolve from real workflow feedback.
Built as MCP Maker delivery 001
Deputy Workforce Operations MCP is the flagship public implementation from MCP Maker: Hyperdrift turns established SaaS APIs into secure, production-ready AI integrations.
The open-source core remains useful on its own. MCP Maker offers managed deployment, security hardening, custom workflows, and operational support for organisations that want a production integration tailored to their environment.
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.