mcp-mayo
MCP server that exposes MAYO Apollo HRM platform's Foundation, Attendance, and Payroll APIs as AI-callable tools, enabling natural language queries for employee profiles, attendance summaries, and payroll reports.
README
mcp-mayo
MCP Server for MAYO Apollo — exposes the HRM platform's Foundation (人事), Attendance (差勤), and Payroll (薪資勞健保) APIs as AI-callable tools via the Model Context Protocol.
繁體中文 · Part of the Asgard AI Platform open-source ecosystem.
Features
- 33 tools — 28 thin wrappers around every MAYO Apollo endpoint plus 5 semantic compositions
- Three backend domains — Foundation, Attendance, Payroll — served from one MCP server with a single
hrmlicenseAPI key - Date-format normalization — callers use ISO
YYYY-MM-DD/YYYY-MM; the server converts to each endpoint's expected format internally - Semantic aggregations — one-call helpers like
get_employee_profile,get_organization_snapshot_as_of,get_attendance_summary - Typed with Pydantic — every parameter has a description the AI can see
- E2E test runner — exercises every registered tool against the live PRE environment
Prerequisites
- Python 3.10+
uv(recommended) orpip- A MAYO-issued
hrmlicenseAPI key with read access to the FD / PT / PY backends you plan to call
Installation
From source (current state)
git clone https://github.com/asgard-ai-platform/mcp-mayo.git
cd mcp-mayo
uv sync
cp .env.example .env
# Edit .env and set MAYO_API_KEY
From PyPI (once published)
uv add mcp-mayo
# or
pip install mcp-mayo
Configuration
| Env var | Required | Purpose |
|---|---|---|
MAYO_API_KEY |
Yes | The token value placed on the hrmlicense header; authenticates Foundation, Attendance, and Payroll in a single credential |
Usage
Run locally
uv run --env-file .env python mcp_server.py
Claude Desktop
{
"mcpServers": {
"mayo": {
"command": "uvx",
"args": ["mcp-mayo"],
"env": {
"MAYO_API_KEY": "your_hrmlicense_token"
}
}
}
}
Claude Code (.mcp.json)
{
"mcpServers": {
"mayo": {
"command": "uv",
"args": ["run", "mcp-mayo"],
"cwd": "${PWD}",
"env": {
"PYTHONPATH": "${PWD}",
"MAYO_API_KEY": "${MAYO_API_KEY}"
}
}
}
}
Cursor / other IDEs
Point the MCP client at uvx mcp-mayo with MAYO_API_KEY in its environment.
Tools
Semantic (recommended for AI use)
| Tool | What it does |
|---|---|
get_employee_profile |
One employee's PA + work history + education in a single call |
get_organization_snapshot_as_of |
Org tree + active headcount for a date |
get_attendance_summary |
History + abnormal + overtime + leave fan-out over a range |
search_active_employees |
list_active_employees with today as default |
get_monthly_payroll_report |
Single-month salary/bonus roster using ISO YYYY-MM |
Foundation (FD) — 15
list_active_employees ¹, list_resigned_employees ¹, get_organization_tree ¹, export_organization_with_employees, export_employees, export_departments, export_organization_full, export_employees_full, export_expatriations, export_working_history, export_education_history, export_employee_changes, export_organization_changes, export_pa_options, export_company_custom_fields
Attendance (PT) — 8
get_attendance_rules, get_attendance_history, get_attendance_abnormal, get_forgot_checkin_records, get_leave_history, get_employee_calendar, get_overtime_records, get_trip_history ²
Payroll (PY) — 5
get_monthly_labor_insurance ², get_monthly_nhi ², get_monthly_labor_pension ², get_salary_insurance_detail, get_salary_bonus_list
¹ The three /ClientOut/ReportCenter/* routes (plus the two semantic wrappers search_active_employees and get_organization_snapshot_as_of that depend on them) currently return 404 on the PRE Foundation backend. The tools stay exposed so they work the moment MAYO redeploys these endpoints.
² Flagged as failing upstream in MAYO's own Postman collection. Same story — tool stays, docstring carries a KNOWN ISSUE note.
Usage Examples
"Who was active in department C030010 yesterday?"
You: 昨天 C030010 部門還在職的人列一下
AI calls:
search_active_employees(
dept_code = "C030010",
)
Result: SUCCESS — returns the active-employee list from the Foundation backend with today as the effective date.
"Give me A00384's full profile"
You: 給我員編 A00384 的完整資料
AI calls:
get_employee_profile(
employee_number = "A00384",
)
Result: SUCCESS — aggregates /PA, /Working, and /Education, locally filtered to A00384.
"Attendance picture for the first week of September"
You: 9/1 到 9/7 的出勤統計,含異常跟加班
AI calls:
get_attendance_summary(
start_date = "2025-09-01",
end_date = "2025-09-07",
)
Result: SUCCESS — returns history, abnormal, overtime, and leave sections for the range.
"January salary roster"
You: 2025-01 的薪資發放清冊
AI calls:
get_monthly_payroll_report(
year_month = "2025-01",
)
Result: SUCCESS — converts to 2025/01 under the hood and calls SalaryBonusList.
Architecture
stdio (JSON-RPC 2.0)
→ mcp_server.py entry point; imports tool modules
→ app.py FastMCP("mcp-mayo") singleton
→ tools/ @mcp.tool() decorated functions
foundation_tools.py 15 FD wrappers
attendance_tools.py 8 PT wrappers
payroll_tools.py 5 PY wrappers
semantic_tools.py 5 high-level compositions
→ connectors/rest_client.py retry + multi-domain URL resolution
→ auth/api_key.py builds {"hrmlicense": <MAYO_API_KEY>}
→ config/settings.py BASE_URLS + ENDPOINTS (keyed by domain)
Testing
uv run --env-file .env python scripts/auth/test_connection.py # probes FD / PT / PY
uv run --env-file .env python tests/test_all_tools.py # runs every tool
Contributing
See CONTRIBUTING.md.
License
MIT — see LICENSE.
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.