ServiceNow MCP Server and API Wrapper

What is ServiceNow MCP Server and API Wrapper about?

Enables developers to access and manipulate ServiceNow data through a clean Python interface. It also includes a standalone MCP server that can be deployed for Agentic AI integrations, exposing ServiceNow functionality over HTTP.

How to use ServiceNow MCP Server and API Wrapper?

1. Install the library: python -m pip install servicenow-api
2. Create a client with either OAuth or basic authentication:

import servicenow_api
client = servicenow_api.Api(
    url="https://your-instance.service-now.com",
    username="YOUR_USERNAME",
    password="YOUR_PASSWORD",
    client_id="YOUR_CLIENT_ID",          # optional for OAuth
    client_secret="YOUR_CLIENT_SECRET"   # optional for OAuth
)

3. Call API methods such as client.get_table(table="incident") or use specialized helpers for change management, CI/CD, etc.
4. Run as an MCP server (Docker example):

docker pull knucklessg1/servicenow:latest

Configure compose.yml and mcp.json as shown in the README, then start the container.

Key features of ServiceNow MCP Server and API Wrapper

• Comprehensive API coverage: Application Service, Change Management, CI/CD, CMDB, Import Sets, Incident, Knowledge Base, generic Table access, and custom endpoints.
• Dual authentication: Supports OAuth 2.0 and basic username/password.
• Proxy & SSL verification options for enterprise network environments.
• Response models with easy attribute access (parent.value.nested_value) and model_dump() for dictionary output.
• MCP server mode: Deployable as a Docker container to expose ServiceNow capabilities to AI agents.
• Extensible: When an endpoint is not pre‑implemented, the generic custom request function allows GET/POST/PUT/DELETE on any ServiceNow REST path.

Use cases of ServiceNow MCP Server and API Wrapper

• Automated ticket handling: AI agents can create, update, or resolve incidents directly via the MCP server.
• Change management workflows: Programmatically trigger, approve, or audit change requests.
• CI/CD integration: Retrieve deployment data from ServiceNow and sync with external pipelines.
• Knowledge base queries: Pull articles for AI‑driven support assistants.
• Custom reporting: Extract CMDB or import‑set data for analytics dashboards.

FAQ from the ServiceNow MCP Server and API Wrapper

Q: Do I need a ServiceNow developer instance? A: Yes, a reachable ServiceNow instance with proper API permissions is required.

Q: Which authentication method should I choose? A: OAuth is recommended for production environments; basic auth works for quick tests or on‑prem setups.

Q: Can I extend the wrapper with my own endpoints? A: Absolutely. Use client.custom_request(method, endpoint, data) to call any ServiceNow REST API not covered out‑of‑the‑box.

Q: How do I run the MCP server without Docker? A: The package can be launched via uv run --with servicenow-api servicenow-mcp as indicated in the mcp.json configuration.

Q: What Python versions are supported? A: The library follows typical packaging practices; Python 3.8+ is recommended.

ServiceNow API

Version: 1.1.4

ServiceNow API Python Wrapper

This repository is actively maintained and will continue adding more API calls

This can run as a standalone MCP Server for Agentic AI!

Contributions are welcome!

All API Response objects are customized for the response call. You can get all return values in a parent.value.nested_value format, or you can run parent.model_dump() to get the table in dictionary format.

API Calls:

• Application Service
• Change Management
• CI/CD
• CMDB
• Import Sets
• Incident
• Knowledge Base
• Table
• Custom Endpoint

If your API call isn't supported, you can always run the standard custom API endpoint function to get/post/put/delete and endpoint

OAuth Authentication

#!/usr/bin/python
# coding: utf-8
import servicenow_api

username = "<SERVICENOW USERNAME>"
password = "<SERVICENOW PASSWORD>"
client_id = "<SERVICENOW CLIENT_ID>"
client_secret = "<SERVICENOW_CLIENT_SECRET>"
servicenow_url = "<SERVICENOW_URL>"

client = servicenow_api.Api(url=servicenow_url,
                            username=username,
                            password=password,
                            client_id=client_id,
                            client_secret=client_secret)

table = client.get_table(table="<TABLE NAME>")
print(f"Table: {table.model_dump()}")

Basic Authentication

#!/usr/bin/python
# coding: utf-8
import servicenow_api

username = "<SERVICENOW USERNAME>"
password = "<SERVICENOW PASSWORD>"
servicenow_url = "<SERVICENOW_URL>"

client = servicenow_api.Api(url=servicenow_url,
                            username=username,
                            password=password)

table = client.get_table(table="<TABLE NAME>")
print(f"Table: {table.model_dump()}")

Proxy and SSL Verify

#!/usr/bin/python
# coding: utf-8
import servicenow_api

username = "<SERVICENOW USERNAME>"
password = "<SERVICENOW PASSWORD>"
servicenow_url = "<SERVICENOW_URL>"

proxy = "https://proxy.net"

client = servicenow_api.Api(url=servicenow_url,
                            username=username,
                            password=password,
                            proxy=proxy,
                            verify=False)

table = client.get_table(table="<TABLE NAME>")
print(f"Table: {table.model_dump()}")

Use with AI

Deploy MCP Server as a Service

docker pull knucklessg1/servicenow:latest

Modify the compose.yml

services:
  servicenow-mcp:
    image: knucklessg1/servicenow:latest
    environment:
      - HOST=0.0.0.0
      - PORT=8004
    ports:
      - 8004:8004

Configure mcp.json

Recommended: Store secrets in environment variables with lookup in JSON file.

For Testing Only: Plain text storage will also work, although not recommended.

{
  "mcpServers": {
    "servicenow": {
      "command": "uv",
      "args": [
        "run",
        "--with",
        "servicenow-api",
        "servicenow-mcp"
      ],
      "env": {
        "SERVICENOW_INSTANCE": "https://www.servicenow.com",
        "SERVICENOW_USERNAME": "user",
        "SERVICENOW_PASSWORD": "pass",
        "SERVICENOW_CLIENT_ID": "client_id",
        "SERVICENOW_CLIENT_SECRET": "client_secret",
        "SERVICENOW_VERIFY": "False"
      },
      "timeout": 200000
    }
  }
}

Install Python Package

python -m pip install servicenow-api

python ./test/test_servicenow_models.py