mcp -logicmonitor

This MCP server provides a LogicMonitor REST API v3 integration for AI-assisted workflows. It exposes a large catalog of tools (around 215) for monitoring data, including core capabilities like alert management, device management, metrics, dashboards, and more, plus extended AI analysis (alert correlation, anomaly detection, baselines), ML/statistical analysis, APM trace tooling, and optional Ansible Automation Platform integrations. To use it, obtain a LogicMonitor API token with appropriate permissions and configure your MCP client to connect through the uvx runner, passing the lm-mcp package as the source. Once connected, you can prompt the agent to perform common LM tasks, query data, trigger analyses, or run remediation workflows via the available 14 workflow prompts and 24 resources. Tools are organized across core monitoring, extended features, AI analysis, ML methods, and platform integrations, enabling both diagnostic and remediation workflows through natural language prompts and structured prompts. Use the provided CLI examples to verify connectivity and test prompts, for example listing configured MCP connections or requesting a prompt-based RCA or alert query.

Prerequisites:

• Python 3.8+ (recommended)
• Internet access to install Python packages

1. Create and activate a Python virtual environment (optional but recommended):

python -m venv venv
source venv/bin/activate  # On Windows use: venv\Scripts\activate

2. Install the LM MCP server package from PyPI:

pip install lm-mcp

3. Verify installation by listing available commands or running a quick syntax check (depends on your client tooling).

4. Prepare your MCP client configuration using the provided mcp_config (see the generated JSON). Common client commands include using uvx with the lm-mcp-server entry point, for example:

# Example: local test using uvx via the lm-mcp package
uvx --from lm-mcp lm-mcp-server

5. Run the MCP server using the configuration you created in your environment, ensuring the required environment variables are set (LM_PORTAL and LM_BEARER_TOKEN).

Note: If you prefer using a container or other environments, adapt the uvx invocation accordingly; this guide assumes a Python-based setup as described in the documentation.

Tips and considerations:

• Ensure LM_PORTAL is set to your LogicMonitor portal hostname (e.g., yourcompany.logicmonitor.com).
• Generate a dedicated LM API token with API access for safe integration; use LM_BEARER_TOKEN as its value.
• The server supports 14 workflow prompts and 24 resources; look for tool descriptions in the MCP client to understand capabilities and parameters.
• If you encounter authentication or connectivity issues, verify token validity, portal URL, and network access to the LM API endpoints.
• The LM MCP server provides extensive AI and ML tooling, including alert correlation, anomaly detection, baselines, trend analysis, and topology-aware scoring; enable or disable features based on your LM plan and security requirements.
• Optional: configure AWX/Ansible-related prompts only if you plan to use the AAP integration (requires AWX URL and token).
• When upgrading, review the release notes for changes to tool availability, API endpoints, or required environment variables.