広告掲載
MCP.so
ログイン

Talos Linux MCP Server

@Nosmoht

リポジトリ

基本情報

カテゴリ

開発者ツール

ライセンス

MIT

トランスポート

stdio

作者

Nosmoht

投稿者

Thomas Krahn

Talos Linux MCP Server について

MCP server for Talos Linux cluster management via the native gRPC API

#talos#mcp#kubernetes

Talos Linux MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "talos-mcp-server": {
      "command": "npx",
      "args": [
        "talos-mcp"
      ]
    }
  }
}

Talos Linux MCP Server のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

上のボタンで一時的なサンドボックスでこのサーバーを実行し、ツール一覧を取得できます。

概要

What is Talos Linux MCP Server?

An MCP server that exposes Talos Linux cluster management to AI agents (Claude Code, OpenAI Codex, and any MCP-compatible client). It provides structured tools that return machine-readable JSON directly from the Talos gRPC API using the same mTLS credentials as talosctl (~/.talos/config). Designed for operators managing Talos clusters.

How to use Talos Linux MCP Server?

Install via npx talos-mcp (no Go required, Linux/macOS, amd64/arm64), download a binary, or build from source with Go 1.21+. Configure via ~/.talos/config or environment variables like TALOSCONFIG and TALOS_CONTEXT. Start as stdio MCP server or enable Streamable HTTP with TALOS_MCP_HTTP_ADDR. Add to your MCP client config (e.g., .mcp.json for Claude Code, claude_desktop_config.json, or .codex/config.toml).

Key features of Talos Linux MCP Server

  • 14 read‑only tools for cluster introspection (logs, health, services, etc.)
  • 8 mutating tools with explicit safety guards (confirm, dry‑run)
  • Safety profiles and environment variables (read‑only mode, allowed nodes/paths)
  • Maintenance‑mode (insecure) operations with certificate fingerprint pinning
  • Upgrade path validation (at most one minor version at a time)
  • Supports Talos Linux v1.9.x through v1.13.x

Use cases of Talos Linux MCP Server

  • Diagnose cluster health and node issues via structured queries
  • Read logs, dmesg, process lists, and files from any node
  • Perform upgrades or rollbacks with validation and confirmation
  • Apply machine config patches or bootstrap new nodes in maintenance mode
  • Inspect etcd cluster status and take backups

よくある質問

1

What Talos versions are supported?

Tested against Talos Linux v1.9.x through v1.13.x. A startup warning appears if the cluster version is outside this range.

2

How does upgrade path validation work?

The `talos_upgrade` tool ensures at most one minor version jump (e.g., v1.11.x → v1.12.x). Upgrades that skip more are rejected unless `TALOS_MCP_SKIP_VERSION_CHECK=true` is set.

3

How to enable maintenance‑mode (insecure) operations?

Set `TALOS_MCP_ENABLE_INSECURE=true` and `TALOS_MCP_INSECURE_ALLOWED_NODES` to a comma‑separated list of permitted maintenance‑mode IPs/CIDRs. Connection MITM can be mitigated with certificate fingerprint pinning.

4

Can I run the server in read‑only mode?

Yes. Set `TALOS_MCP_READ_ONLY=true` to disable all mutating tools at startup.

5

What transport and authentication is supported?

The server can run over stdio (default) or HTTP (when `TALOS_MCP_HTTP_ADDR` is set). HTTP mode requires a bearer token

コメント

「開発者ツール」の他のコンテンツ

Nebelus

Nebelus

Nebelus is an enterprise agentic-AI platform for regulated industries. This MCP server exposes the Nebelus Construction API: ~48 tools that give an MCP client full agent build parity — create/update/probe agents, edit ag

Casefile

azimov777

One case file per task — shared by every session, agent and person that touches it. Self-hosted MCP server and live board. MIT.

3

Capgo MCP

Cap-go

Console, Backend and CLI to manage Capgo Instant update and Native build for Capacitor apps Build and upload to app store your mobile app

LastPing

LastPing

Know when your AI agents, cron jobs and CI pipelines fail, stall or go quiet. Your agent sets up monitors from a plain description, tells you what is broken and why, reads run history and traces, and writes notes on inci

FARPY

mangomunchr

Verified distributed compute for humans, applications, and agents. Submit Blender, Octane, and video workloads, route them across qualified GPUs, track asynchronous execution, download finished artifacts, and verify FARP

1
広告掲載