Ansible MCP Server

Advanced Ansible Model Context Protocol (MCP) server in Python exposing Ansible utilities for inventories, playbooks, roles, and project workflows with troubleshooting capabilities.

Features

• 36 Comprehensive Tools across Core Ansible, Inventory Management, and Troubleshooting

• Self-Healing Capabilities with automated problem resolution

• Health Monitoring with intelligent scoring and recommendations

• Security Auditing with vulnerability assessment

• Performance Benchmarking and baseline management

• Advanced Log Analysis with pattern recognition and correlation

• Pure Ansible Integration - uses native Ansible modules and commands

Related MCP server: AAP MCP Server

Quick Start with uv

Prerequisites

• Python 3.10 or higher

• uv package manager

• Ansible (ansible-core >= 2.16.0)

• macOS/Linux

Installation

1. Install uv (if not already installed):

curl -LsSf https://astral.sh/uv/install.sh | sh

2. Clone the repository:

git clone https://github.com/vsl8/ansible-mcp-server.git
cd ansible-mcp-server

3. Install dependencies using uv:

# Create virtual environment and install dependencies
uv sync

# Activate the virtual environment
source .venv/bin/activate  # On Linux/macOS
# or
.venv\Scripts\activate     # On Windows

4. Verify installation:

# Test that server can start
uv run python src/ansible_mcp/server.py --help

Configuration

Transport Options

The Ansible MCP Server supports two transport modes:

1. stdio (Local) - For same-machine usage (default)

   • Fastest performance

   • Ideal for local development

   • No network configuration needed

2. SSE (Remote) - For remote Ansible servers

   • Connect to remote Ansible installations

   • HTTP-based Server-Sent Events protocol

   • Supports multiple simultaneous clients

   • See Remote Setup Guide for detailed configuration

Quick Remote Setup

# On remote Ansible server
./setup-remote.sh

# Or manually
uv run python src/ansible_mcp/server.py --transport sse --host 0.0.0.0 --port 8000

For Visual Studio Code (VS Code)

Local Configuration (stdio)

VS Code uses the same MCP configuration as Cursor. Add to ~/.vscode/mcp.json or your workspace settings:

{
  "mcpServers": {
    "ansible-mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "/absolute/path/to/ansible-mcp-server",
        "run",
        "python",
        "src/ansible_mcp/server.py"
      ],
      "env": {
        "MCP_ANSIBLE_PROJECT_ROOT": "/path/to/your/ansible/project",
        "MCP_ANSIBLE_INVENTORY": "/path/to/your/ansible/project/inventory/hosts.ini",
        "MCP_ANSIBLE_PROJECT_NAME": "my-project"
      }
    }
  }
}

Note: Ensure you have the MCP extension installed in VS Code. Search for "Model Context Protocol" in the VS Code Extensions marketplace.

Remote Configuration (SSE)

For connecting to a remote Ansible server:

{
  "mcpServers": {
    "ansible-remote": {
      "url": "http://your-ansible-server:8000/sse",
      "transport": "sse"
    }
  }
}

For Cursor IDE

Local Configuration (stdio)

Add to your Cursor MCP config file (~/.cursor/mcp.json or <project>/.cursor/mcp.json):

{
  "mcpServers": {
    "ansible-mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "/absolute/path/to/ansible-mcp-server",
        "run",
        "python",
        "src/ansible_mcp/server.py"
      ],
      "env": {
        "MCP_ANSIBLE_PROJECT_ROOT": "/path/to/your/ansible/project",
        "MCP_ANSIBLE_INVENTORY": "/path/to/your/ansible/project/inventory/hosts.ini",
        "MCP_ANSIBLE_PROJECT_NAME": "my-project"
      }
    }
  }
}

Remote Configuration (SSE)

For connecting to a remote Ansible server:

{
  "mcpServers": {
    "ansible-remote": {
      "url": "http://your-ansible-server:8000/sse",
      "transport": "sse"
    }
  }
}

For Claude Desktop

Local Configuration (stdio)

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%/Claude/claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "ansible-mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "/absolute/path/to/ansible-mcp-server",
        "run",
        "python",
        "src/ansible_mcp/server.py"
      ],
      "env": {
        "MCP_ANSIBLE_PROJECT_ROOT": "/path/to/your/ansible/project",
        "MCP_ANSIBLE_INVENTORY": "/path/to/your/ansible/project/inventory/hosts.ini",
        "MCP_ANSIBLE_PROJECT_NAME": "my-project"
      }
    }
  }
}

Remote Configuration (SSE)

For connecting to a remote Ansible server:

{
  "mcpServers": {
    "ansible-remote": {
      "url": "http://your-ansible-server:8000/sse",
      "transport": "sse"
    }
  }
}

For GitHub Copilot CLI

Local Configuration (stdio)

GitHub Copilot CLI supports MCP servers. Configure in ~/.github-copilot/config.json:

{
  "mcp": {
    "servers": {
      "ansible-mcp": {
        "command": "uv",
        "args": [
          "--directory",
          "/absolute/path/to/ansible-mcp-server",
          "run",
          "python",
          "src/ansible_mcp/server.py"
        ],
        "env": {
          "MCP_ANSIBLE_PROJECT_ROOT": "/path/to/your/ansible/project",
          "MCP_ANSIBLE_INVENTORY": "/path/to/your/ansible/project/inventory/hosts.ini",
          "MCP_ANSIBLE_PROJECT_NAME": "my-project"
        }
      }
    }
  }
}

Usage with GitHub Copilot CLI:

# Initialize if config doesn't exist
mkdir -p ~/.github-copilot

# Test the MCP server
gh copilot explain "How do I list inventory hosts?"

# Use with GitHub Copilot CLI
gh copilot suggest "Run ansible playbook on production servers"

For Claude CLI

Configure Claude CLI to use the MCP server. Add to ~/.config/claude/config.json:

{
  "mcpServers": {
    "ansible-mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "/absolute/path/to/ansible-mcp-server",
        "run",
        "python",
        "src/ansible_mcp/server.py"
      ],
      "env": {
        "MCP_ANSIBLE_PROJECT_ROOT": "/path/to/your/ansible/project",
        "MCP_ANSIBLE_INVENTORY": "/path/to/your/ansible/project/inventory/hosts.ini",
        "MCP_ANSIBLE_PROJECT_NAME": "my-project"
      }
    }
  }
}

Usage with Claude CLI:

# Install Claude CLI (if not installed)
npm install -g @anthropic-ai/claude-cli

# Initialize configuration
claude config init

# Use with Claude CLI
claude chat "List all ansible hosts in my inventory"

# Run specific MCP tool
claude mcp ansible-mcp ansible-inventory

Environment Variables (Optional)

• MCP_ANSIBLE_PROJECT_ROOT: Absolute path to your Ansible project root

• MCP_ANSIBLE_INVENTORY: Path to inventory file or directory

• MCP_ANSIBLE_PROJECT_NAME: Label for your project

• MCP_ANSIBLE_ROLES_PATH: Colon-separated roles paths

• MCP_ANSIBLE_COLLECTIONS_PATHS: Colon-separated collections paths

• MCP_ANSIBLE_ENV_<KEY>: Forwarded to process env (e.g., MCP_ANSIBLE_ENV_ANSIBLE_CONFIG)

Available Tools

Core Ansible Tools (17)

1. create-playbook - Create playbooks from YAML strings or dicts

2. validate-playbook - Validate playbook syntax

3. ansible-playbook - Execute playbooks

4. ansible-task - Run ad-hoc tasks

5. ansible-role - Execute roles via temporary playbook

6. create-role-structure - Scaffold role directory tree

7. ansible-inventory - List inventory hosts and groups

8. register-project - Register Ansible project for reuse

9. list-projects - Show registered projects

10. project-playbooks - Discover playbooks in project

11. project-run-playbook - Run playbook using project config

12. ansible-ping - Ping hosts

13. ansible-gather-facts - Gather and return facts

14. validate-yaml - Validate YAML files

15. galaxy-install - Install roles/collections

16. galaxy-lock - Generate lock file

17. project-bootstrap - Bootstrap project environment

Inventory Management Suite (6)

1. inventory-parse - Parse inventories with group_vars/host_vars

2. inventory-graph - Show inventory graph

3. inventory-find-host - Find host's groups and variables

4. inventory-diff - Compare two inventories

5. ansible-test-idempotence - Test playbook idempotence

6. vault-* - Vault operations (encrypt, decrypt, view, rekey)

Advanced Troubleshooting Suite (13)

Foundation Tools (3)

• ansible-remote-command - Execute shell commands with parsing

• ansible-fetch-logs - Fetch and analyze log files

• ansible-service-manager - Manage services with logs

Intelligent Diagnostics (3)

• ansible-diagnose-host - Comprehensive health assessment

• ansible-capture-baseline - Capture system state snapshots

• ansible-compare-states - Compare against baselines

Automation (1)

• ansible-auto-heal - Automated problem resolution

Network & Security (2)

• ansible-network-matrix - Network connectivity testing

• ansible-security-audit - Security vulnerability assessment

Performance & Monitoring (4)

• ansible-health-monitor - Continuous monitoring with trends

• ansible-performance-baseline - Performance benchmarking

• ansible-log-hunter - Advanced log correlation

Development

Using uv for Development

# Install with dev dependencies
uv sync --all-extras

# Run tests
uv run pytest

# Type checking
uv run mypy src/

# Linting
uv run ruff check src/

# Format code
uv run ruff format src/

Project Structure

ansible-mcp-server/
├── src/
│   └── ansible_mcp/
│       ├── __init__.py
│       ├── server.py          # Main MCP server (2247 lines, 36 tools)
│       └── py.typed
├── pyproject.toml             # Project configuration (uv-compatible)
├── .python-version            # Python version
├── .gitignore
└── README.md

Usage Examples

List Inventory Hosts

# Tool: ansible-inventory
# Args: 
{
  "inventory": "/path/to/inventory/hosts.ini"
}

Run a Playbook

# Tool: ansible-playbook
# Args:
{
  "playbook_path": "/path/to/playbook.yml",
  "inventory": "/path/to/inventory",
  "extra_vars": {"env": "production"}
}

Health Check with Recommendations

# Tool: ansible-diagnose-host
# Args:
{
  "host_pattern": "webservers",
  "checks": ["system", "network", "security", "performance"],
  "include_recommendations": true
}

Automated Healing (Dry Run)

# Tool: ansible-auto-heal
# Args:
{
  "host_pattern": "database",
  "symptoms": ["high_memory", "disk_full"],
  "max_impact": "medium",
  "dry_run": true
}

Network Connectivity Matrix

# Tool: ansible-network-matrix
# Args:
{
  "host_patterns": ["web*", "db*"],
  "check_ports": [22, 3306, 443]
}

Advantages of Using uv

✅ Fast: 10-100x faster than pip
✅ Reliable: Consistent dependency resolution
✅ Simple: Single tool for all Python package management
✅ Compatible: Works with existing pyproject.toml
✅ Lockfile: Automatic lock file generation for reproducibility

Troubleshooting

uv not found

# Install uv
curl -LsSf https://astral.sh/uv/install.sh | sh
# Add to PATH if needed
export PATH="$HOME/.local/bin:$PATH"

Python version mismatch

# Install specific Python version with uv
uv python install 3.10

# Pin Python version for project
uv python pin 3.10

Dependencies not installing

# Clean and reinstall
rm -rf .venv
uv sync --refresh

MCP server not starting

# Test the server directly
uv run python src/ansible_mcp/server.py

# Check logs (stderr output)
uv run python src/ansible_mcp/server.py 2> server.log

Requirements

• Python 3.10+

• uv >= 0.1.0

• mcp[cli] >= 1.2.0

• ansible-core >= 2.16.0

• PyYAML >= 6.0.1

License

MIT License - See LICENSE file for details

Contributing

Contributions are welcome! Please:

1. Fork the repository

2. Create a feature branch

3. Make your changes

4. Add tests if applicable

5. Run linters and tests with uv

6. Submit a pull request

Acknowledgments

• Based on the excellent work by bsahane/mcp-ansible

• Built with FastMCP

• Powered by Ansible

Support

For issues, questions, or contributions:

• GitHub Issues: https://github.com/vsl8/ansible-mcp-server/issues

• Documentation: https://github.com/vsl8/ansible-mcp-server#readme

Note: This server uses stdio transport. Do not print to stdout; logs go to stderr.-server