MCP Catalogs
Homelsbot screenshot

lsbot

by ruilisi·409·Score 50

A lean, secure bot with MCP protocol integration and end-to-end encryption for AI applications.

developer-toolssecurityai-llmcommunication
57
Forks
3
Open issues
1 mo ago
Last commit
2d ago
Indexed

Overview

lsbot is a Lean & Secure Bot built in Go that prioritizes user privacy through end-to-end encryption. It implements the Model Context Protocol (MCP) to expose 75+ tools for file system operations, shell commands, system information, process management, networking, calendar, reminders, browser automation, and more. Unlike similar solutions, lsbot keeps all data locally and encrypts communication with relay servers, ensuring sensitive information remains on the user's machine. The project supports multiple AI providers including Claude, DeepSeek, Kimi, and offers gateway modes for various messaging platforms like WeChat Work, Telegram, and Discord.

Try asking AI

After installing, here are 7 things you can ask your AI assistant:

you:Secure AI assistant with local data processing and encrypted communication
you:Automated workflow integration with messaging platforms like WeChat Work and Slack
you:Browser automation without Node.js dependencies using CDP protocol
you:Cross-platform file management and system control through AI agents
you:How is lsbot different from other AI bots?
you:Which AI providers are supported?
you:How does the MCP protocol integration work?

When to choose this

Choose lsbot when you need a secure AI bot for enterprise messaging platforms with end-to-end encryption and comprehensive system access via MCP tools.

When NOT to choose this

Avoid lsbot if you need extensive third-party integrations beyond its supported platforms or require cloud-based conversation history and data storage.

Tools this server exposes

12 tools extracted from the README
  • query_postgres

    Execute SQL queries on PostgreSQL databases

  • list_files

    List files and directories in a specified path

  • execute_command

    Execute shell commands and return their output

  • create_github_issue

    Create a new issue on GitHub repository

  • read_file

    Read and return the contents of a file

  • send_notification

    Send system notifications to the user

  • search_web

    Perform web searches using DuckDuckGo

  • take_screenshot

    Capture screenshots of the screen

  • git_commit

    Create git commits with automatic message generation

  • run_browser_automation

    Automate browser actions using Chrome DevTools Protocol

  • schedule_task

    Create and manage scheduled cron jobs

  • get_weather

    Fetch current weather and forecasts

Comparable tools

openclawclaude-desktopcursorwindsurf

Installation

# macOS / Linux / WSL
curl -fsSL https://files.lingti.com/install-lsbot.sh | bash

# Windows (PowerShell)
irm https://files.lingti.com/install-lsbot.ps1 | iex

For Claude Desktop integration, add this to Claude Desktop's config.json:

{
  "mcpServers": {
    "lsbot": {
      "command": "lsbot",
      "args": ["mcp"]
    }
  }
}

FAQ

How is lsbot different from other AI bots?
lsbot prioritizes security with end-to-end encryption, local data storage, no runtime dependencies, and a single binary executable. Unlike many alternatives, it doesn't route messages through developer servers that could read user data.
Which AI providers are supported?
lsbot supports 16+ AI providers including Claude, DeepSeek, Kimi, MiniMax, Gemini, OpenAI, and others. You can configure different providers for different agents or channels.
How does the MCP protocol integration work?
lsbot implements the full MCP protocol, exposing its tools and resources to MCP-compatible clients like Claude Desktop, Cursor, and Windsurf. This allows AI clients to access local system resources securely through lsbot.

Compare lsbot with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.