sanshu

🥈Silver

Sanshu is an open-source binary MCP that automates operations tasks. It connects to Claude agents to streamline workflows, reducing manual effort and improving efficiency for operations teams.

216140Updated 3d ago

Intermediate30min to implementautomation

Saves ~300 min per use

Quick InstallView Source

git clone https://github.com/yuaotian/sanshu.git

Works with:

Claude

Overview

About This Skill

Sanshu is an open-source binary MCP that automates operations tasks. It connects to Claude agents to streamline workflows, reducing manual effort and improving efficiency for operations teams.

How to Use

["Install the Sanshu MCP server in your Claude environment using the MCP CLI: `mcp install sanshu`","Configure the server with your target environment credentials (e.g., AWS, Kubernetes, or local system access).","Craft a prompt using the template above, replacing [OPERATION_TASK], [SYSTEM_OR_ENVIRONMENT], and [PARAMETERS_IF_ANY] with your specific needs.","Run the prompt in your AI assistant and review the automation report. Adjust parameters if needed and re-run.","For complex workflows, chain multiple Sanshu tasks together using Claude's multi-step reasoning or create a custom script that calls Sanshu operations sequentially."]

Setup & Installation

Quick Install

No install command available. Check the GitHub repository for manual installation instructions.

Alternative Install (Git Clone)

git clone https://github.com/yuaotian/sanshu

Requirements

Claude Code or compatible AI agent
Works with: Claude

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

Use the Sanshu MCP server to automate [OPERATION_TASK] for [SYSTEM_OR_ENVIRONMENT]. Include the following parameters: [PARAMETERS_IF_ANY]. Execute the task and provide a summary of the actions taken, results, and any errors encountered.

Example Output

### Sanshu MCP Automation Report
**Task:** Restart failed system services on the production web tier
**Environment:** AWS EC2 (us-west-2), Ubuntu 22.04 LTS
**Parameters:** --service nginx,postgresql,redis --timeout 30s --notify slack_channel #ops-alerts

**Actions Executed:**
1. **Service Discovery:** Scanned systemd for services matching nginx, postgresql, or redis (found 3 active services)
2. **Status Check:** Verified current status of each service:
   - nginx: active (running)
   - postgresql: inactive (dead) [FAILED]
   - redis: active (running)
3. **Restart Attempt:** Issued `systemctl restart postgresql` (attempt 1/3)
4. **Verification:** Confirmed postgresql status changed to active (running)
5. **Notification:** Sent Slack alert to #ops-alerts with restart confirmation

**Results:**
- Successfully restarted 1 service (postgresql)
- 2 services required no action (nginx, redis)
- Total execution time: 8.4 seconds
- No errors encountered

**Recommendation:** Monitor postgresql logs for 15 minutes to ensure stability. Schedule a weekly health check for this tier.