IT Onboarding Automator MCP Server
Enables AI assistants to inspect employee access, list failed onboarding events, and retry provisioning operations.
README
Mock IT Onboarding Automator
Project Overview
The Mock IT Onboarding Automator is a backend service that automates employee onboarding based on HR webhook events.
When a new employee is hired, the HR system sends an onboarding event to the service. The system validates the event, determines which applications the employee should receive based on their role, provisions access records, records audit information, and ensures duplicate events are handled safely through idempotent processing.
The project also exposes an MCP (Model Context Protocol) server that allows AI assistants and operators to inspect employee access, review failed onboarding events, and retry failed provisioning operations.
Features
Webhook Processing
Supports HR onboarding events via:
POST /webhooks/hris
Supported event type:
employee.hired
Role-Based Access Provisioning
Application access is automatically assigned based on employee role.
engineer
- slack
- google_workspace
- jira
sales
- slack
- google_workspace
- salesforce
it_admin
- slack
- google_workspace
- jira
- salesforce
Idempotent Event Processing
Duplicate webhook deliveries using the same event_id are safely ignored.
The system guarantees:
- No duplicate employees
- No duplicate access grants
- No duplicate provisioning actions
Audit Logging
Every successful provisioning operation creates an audit log containing:
- event_id
- role
- granted applications
- idempotency status
MCP Server
The MCP server exposes the following tools:
get_employee_access
Retrieve employee details and provisioned application access.
list_failed_events
List failed onboarding events.
retry_provision
Retry a previously failed onboarding event.
Architecture Summary
flowchart TB
HR[HR System]
HR -->|POST /webhooks/hris| WEBHOOK[FastAPI Webhook]
WEBHOOK --> PROV[Provisioning Service]
PROV --> EMP[Employees]
PROV --> ACCESS[Access Grants]
PROV --> AUDIT[Audit Log]
DB[(SQLite Database)]
EMP --> DB
ACCESS --> DB
AUDIT --> DB
MCP[MCP Server]
MCP --> TOOL1[get_employee_access]
MCP --> TOOL2[list_failed_events]
MCP --> TOOL3[retry_provision]
TOOL1 --> DB
TOOL2 --> DB
TOOL3 --> DB
Technology Stack
| Component | Technology |
|---|---|
| Language | Python 3.12+ |
| API Framework | FastAPI |
| ORM | SQLAlchemy 2.0 |
| Database | SQLite |
| MCP | Official Python MCP SDK |
| Testing | pytest |
| Package Management | uv |
Prerequisites
Install:
- Python 3.12+
- uv
- Git
Verify installation:
python --version
uv --version
git --version
Installation
Clone repository:
git clone https://github.com/Aman3786/IT-Onboarding-Automator.git
cd IT-Onboarding-Automator
Install dependencies:
uv sync
OR
uv pip install -r requirements.txt
Initialize Database
Create tables and seed initial role mappings:
uv run python setup_db.py
Expected output:
Database initialized successfully
Database location:
data/onboarding.db
Run API
Start FastAPI server:
uv run uvicorn api.main:app --reload
API available at:
http://localhost:8000
Interactive documentation:
http://localhost:8000/docs
Run MCP Server
Start MCP server (Prequisite: Nodejs Should be installed for MCP Inspector)
npx @modelcontextprotocol/inspector uv run python -m mcp_server.server
OR
uv run mcp dev mcp_server/server.py
OR
uv run python -m mcp_server.server
npx -y @modelcontextprotocol/inspector
The MCP server uses stdio transport and is intended to be consumed by MCP Inspector, cursor and other MCP-compatible clients.
Configure Cursor MCP
Create:
.cursor/mcp.json
Configuration:
{
"mcpServers": {
"onboarding-automator": {
"command": "uv",
"args": [
"run",
"python",
"-m",
"mcp_server.server"
]
}
}
}
Restart Cursor after creating the configuration.
The following tools should become available through MCP Inspector/Cursor:
- get_employee_access
- list_failed_events
- retry_provision
Run Tests
Run all tests:
uv run pytest
Run verbose output:
uv run pytest -v
Run a specific test file:
uv run pytest tests/test_webhook.py -v
Example Requests
Successful Employee Onboarding
curl -X POST http://localhost:8000/webhooks/hris \
-H "Content-Type: application/json" \
-d '{
"event_id":"evt_hire_001",
"event_type":"employee.hired",
"email":"alex.chen@example.com",
"full_name":"Alex Chen",
"role":"engineer"
}'
Example response:
{
"event_id": "evt_hire_001",
"status": "completed",
"idempotent": false,
"employee": {
"email": "alex.chen@example.com",
"role": "engineer"
},
"granted_apps": [
"slack",
"google_workspace",
"jira"
]
}
Duplicate Event
Submitting the same request again:
{
"event_id": "evt_hire_001",
"status": "completed",
"idempotent": true
}
Invalid Role
curl -X POST http://localhost:8000/webhooks/hris \
-H "Content-Type: application/json" \
-d '{
"event_id":"evt_invalid_role",
"event_type":"employee.hired",
"email":"bad@example.com",
"full_name":"Bad User",
"role":"unknown_role"
}'
Response:
{
"event_id":"evt_invalid_role",
"status":"failed",
"error":"Unknown role: 'unknown_role'"
}
Assumptions
- Employee email addresses are unique.
- Roles are predefined and managed internally.
- Application provisioning is simulated through database records.
- SQLite is sufficient for local execution and evaluation.
- Duplicate webhook deliveries reuse the same event_id.
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.