talkops
This repository will contains code specific to all TalkOps MCP servers associated with various DevOps tools and technologies.
claude mcp add --transport stdio talkops-ai-talkops-mcp docker run -p 8765:8765 -v ~/.kube/config:/app/.kube/config:ro talkopsai/helm-mcp-server:latest
How to use
TalkOps serves as a centralized MCP registry and launcher for a suite of DevOps MCP servers. It exposes multiple domain-specific servers (e.g., Helm for Kubernetes package management, ArgoCD for GitOps, Terraform for IaC, and an Agents Registry) that AI agents can interact with through a uniform interface. Each server provides capabilities such as chart operations, application sync, infrastructure as code execution, credential isolation, and discovery of available agent capabilities. To get started, run the individual MCP servers in your environment using the recommended Quick Install commands, then query the server endpoints or use generated MCP clients to send context-aware requests (e.g., search charts, apply manifests, run Terraform plans, or discover agents). The ecosystem is designed to be modular, so you can deploy only the servers you need and connect them through the TalkOps registry as your automation requirements grow.
How to install
Prerequisites:
- Docker (or Python uv/uvx runtime for non-Docker deployments)
- Access to your cluster kubeconfig if using Kubernetes tools
- Optional: SSH keys for accessing Git repositories or remote servers
Step-by-step installation (using Quick Install commands from the README):
- Install and run Helm MCP Server (Docker): docker run -p 8765:8765 -v ~/.kube/config:/app/.kube/config:ro talkopsai/helm-mcp-server:latest
- Install and run ArgoCD MCP Server (Docker):
docker run -p 8765:8765 -e ARGOCD_SERVER_URL=... -e ARGOCD_AUTH_TOKEN=... -v ~/.ssh/id_ed25519:/app/.ssh/id_rsa:ro
sandeep2014/talkops-mcp:argocd-mcp-server-latest - Install Terraform MCP Server (Python/uv micro-runtime): cd src/terraform-mcp-server uv pip install -e . uv run agents-mcp-server
- Install Agents Central Registry (Python/uv micro-runtime): cd src/agents-mcp-server uv pip install -e . uv run -m agents_mcp_server Notes:
- Adjust environment variable values (e.g., ARGOCD_SERVER_URL, ARGOCD_AUTH_TOKEN) to your environment.
- If you don’t use Docker, follow the repository README sections for UV/UVX-based installations for the Python servers.
Additional notes
Tips and common issues:
- Ensure network access between the host and containers if Kubernetes config or SSH keys are mounted.
- For ArgoCD, use credential isolation to avoid leaking tokens; prefer ephemeral tokens or secrets management.
- The Terraform MCP Server supports multi-provider backends; ensure credentials for AWS/Azure/GCP are properly configured in the environment or secret store.
- When using the Agents Registry, keep the registry up to date to reflect newly added agents and capabilities.
- If you run into port conflicts, change the host port mappings in the Quick Install commands and update any client configurations accordingly.
Related MCP Servers
nautex
MCP server for guiding Coding Agents via end-to-end requirements to implementation plan pipeline
mcp-yfinance
Real-time stock API with Python, MCP server example, yfinance stock analysis dashboard
cloudwatch-logs
MCP server from serkanh/cloudwatch-logs-mcp
servicenow-api
ServiceNow MCP Server and API Wrapper
the -company
TheMCPCompany: Creating General-purpose Agents with Task-specific Tools
mcp-yahoo-finance
MCP server for Yahoo Finance