litmus
MCP server from litmuschaos/litmus-mcp-server
claude mcp add --transport stdio litmuschaos-litmus-mcp-server docker run -i litmuschaos-mcp-server:latest \ --env DEFAULT_INFRA_ID="optional-default-infra-id" \ --env LITMUS_PROJECT_ID="your-project-id" \ --env LITMUS_ACCESS_TOKEN="your-token" \ --env CHAOS_CENTER_ENDPOINT="http://your-chaos-center:8080" \ --env DEFAULT_ENVIRONMENT_ID="production"
How to use
LitmusChaos MCP Server is a Go-based MCP server that provides a comprehensive control plane for managing LitmusChaos resources through AI-assisted interactions. It exposes a rich set of tools to manage chaos experiments, infrastructures, environments, resilience probes, chaos hubs, and analytics. With this MCP server, you can list, describe, run, stop, and monitor experiments; manage infrastructures and environments; create resilience probes; browse ChaosHubs and faults; and fetch statistics and run details. The server is designed to integrate with AI assistants or vault-like assistants (e.g., Claude) to enable natural-language driven chaos engineering workflows. Typical usage flows include listing available experiments, triggering on-demand experiments, monitoring live runs, and querying for historical results and resiliency scores. Tools are grouped by areas like Experiment Management, Execution Monitoring, Infrastructure Management, Environment Organization, Resilience Validation, and Discovery & Analytics, enabling precise natural-language queries such as “Show me all production experiments with status running” or “Create an HTTP resilience probe for the payment API.”
How to install
Prerequisites:
- Go 1.21 or higher (for building from source)
- Docker (optional, for containerized deployment)
- Access to a LitmusChaos 3.x Chaos Center and valid credentials
From Source
-
Clone the repository git clone https://github.com/yourusername/litmuschaos-mcp-server.git cd litmuschaos-mcp-server
-
Build the binary make build
-
Install locally make install
Using Go Install
- Install the latest release go install github.com/yourusername/litmuschaos-mcp-server@latest
Using Docker
-
Build the Docker image (if a Dockerfile is provided in repo) make docker-build
-
Run with required environment variables docker run --rm -it
-e CHAOS_CENTER_ENDPOINT=http://your-chaos-center:8080
-e LITMUS_PROJECT_ID=your-project-id
-e LITMUS_ACCESS_TOKEN=your-token
litmuschaos-mcp-server:latest
Additional notes
Tips and common issues:
- Ensure CHAOS_CENTER_ENDPOINT, LITMUS_PROJECT_ID, and LITMUS_ACCESS_TOKEN are correctly set; missing or invalid credentials will prevent API access.
- When using Docker, prefer mounting environment-specific values and consider using a secret manager for tokens.
- If you encounter connectivity issues, verify Chaos Center is reachable from the MCP server host and that firewalls allow access.
- For development workflows, the repository supports hot-reload development setup with Makefile targets (dev, build, test, etc.).
- The MCP server supports multiple environments and can filter operations by environment or namespace depending on setup; use DEFAULT_INFRA_ID and DEFAULT_ENVIRONMENT_ID to simplify common commands.
- When using Claude or other assistants, ensure the MCP server endpoint in your client configuration points to the MCP server (e.g., http://localhost:8080) and that the environment variables align with your deployment.
Related MCP Servers
trpc-agent-go
trpc-agent-go is a powerful Go framework for building intelligent agent systems using large language models (LLMs) and tools.
station
Station is our open-source runtime that lets teams deploy agents on their own infrastructure with full control.
tiger-cli
Tiger CLI is the command-line interface for Tiger Cloud. It includes an MCP server for helping coding agents write production-level Postgres code.
gopls
MCP server for golang projects development: Expand AI Code Agent ability boundary to have a semantic understanding and determinisic information for golang projects.
kubernetes
A Model Context Protocol (MCP) server for the Kubernetes API.
gcp-cost
💰 An MCP server that enables AI assistants to estimate Google Cloud costs, powered by Cloud Billing Catalog API and built with Genkit for Go