List incidents from Rootly with pagination (default 10 items)
Get details of a specific incident by ID
List alerts associated with a specific incident
Rootly MCP Server provides seamless integration with Rootly's incident management platform directly within your favorite MCP-compatible code editors like Cursor, Windsurf, and Claude. This integration allows developers and SREs to view, manage, and resolve production incidents without context switching between tools. The server dynamically generates MCP resources based on Rootly's OpenAPI specification, providing access to critical incident data with sensible defaults like pagination to prevent context window overflow. By bringing incident management capabilities directly into your development environment, Rootly helps teams resolve production issues faster and more efficiently.
uv
package manager (install with curl -LsSf https://astral.sh/uv/install.sh | sh
)The simplest way to install Rootly MCP Server is through PyPI:
pip install rootly-mcp-server
For more customization options, you can clone the repository:
git clone https://github.com/Rootly-AI-Labs/Rootly-MCP-server.git
cd Rootly-MCP-server
uv pip install -e .
To set up Rootly MCP Server in your MCP-compatible editor (such as Cursor or Windsurf), add the following configuration:
{
"mcpServers": {
"rootly": {
"command": "uvx",
"args": [
"--from",
"rootly-mcp-server",
"rootly-mcp-server"
],
"env": {
"ROOTLY_API_TOKEN": "<YOUR_ROOTLY_API_TOKEN>"
}
}
}
}
If you want to customize the allowed API paths or other settings, use this configuration after cloning the repository:
{
"mcpServers": {
"rootly": {
"command": "uv",
"args": [
"run",
"--directory",
"/path/to/rootly-mcp-server",
"rootly-mcp-server"
],
"env": {
"ROOTLY_API_TOKEN": "<YOUR_ROOTLY_API_TOKEN>"
}
}
}
}
Once configured, you can interact with Rootly directly from your AI-assisted editor. For example, you can:
The server is configured with sensible defaults:
By default, the server only exposes the /incidents
and /incidents/{incident_id}/alerts
endpoints from Rootly's API. This limitation helps prevent context overflow and improves AI agent performance.
If you need access to additional Rootly API endpoints, you can modify the allowed_paths
variable in src/rootly_mcp_server/server.py
after cloning the repository.
If you encounter issues:
uv
package manager is properly installedFor more help, visit the Rootly AI Labs GitHub repository or submit an issue.