
win4r/openclaw-remote-minimax-setup-skill
Developer toolsReusable OpenClaw skill for remote Linux deployment with MiniMax M2.1 and Telegram bot setup
Overview
A reusable OpenClaw skill for deploying and validating with: - - - - real-world first-boot troubleshooting This repository packages the exact deployment pattern distilled from a successful live setup, but all secrets and host-specific details have been removed. The skill is designed for agents or operators who need a repeatable workflow for: 1. reading the official OpenClaw docs first 2. installing OpenClaw safely on a remote server 3. configuring minimax/MiniMax-M2.1 4. wiring up a Telegram bot 5. validating gateway.mode=local 6. fixing the classic systemctl --user persistence issue with loginctl enable-linger 7. verifying both the model path and the Telegram path This skill explicitly captures several deployment pitfalls that are easy to miss: - openclaw config validate can pass while the gateway still refuses first boot - missing gateway.
README
OpenClaw Remote MiniMax Setup Skill
A reusable OpenClaw skill for deploying and validating OpenClaw on a remote Linux server over SSH with:
- MiniMax M2.1
- MiniMax China API endpoint
- Telegram bot integration
- real-world first-boot troubleshooting
This repository packages the exact deployment pattern distilled from a successful live setup, but all secrets and host-specific details have been removed.
What this skill covers
The skill is designed for agents or operators who need a repeatable workflow for:
- reading the official OpenClaw docs first
- installing OpenClaw safely on a remote server
- configuring
minimax/MiniMax-M2.1 - wiring up a Telegram bot
- validating
gateway.mode=local - fixing the classic
systemctl --userpersistence issue withloginctl enable-linger - verifying both the model path and the Telegram path
Repository layout
.
├── README.md
├── README_CN.md
├── dist/
│ └── openclaw-remote-minimax-setup.skill
└── skill/
└── openclaw-remote-minimax-setup/
├── SKILL.md
└── references/
├── config-template.md
├── failure-modes.md
└── workflow.md
Included troubleshooting knowledge
This skill explicitly captures several deployment pitfalls that are easy to miss:
openclaw config validatecan pass while the gateway still refuses first boot- missing
gateway.mode=localcan block startup - Telegram may return
chat not founduntil the user presses Start in the bot DM - a remote
systemctl --userservice can silently die after SSH disconnect unless linger is enabled
How to use
Option 1: Use the source skill folder
Copy the folder below into your OpenClaw skills directory or your own skill workspace:
skill/openclaw-remote-minimax-setup/
Option 2: Use the packaged .skill archive
Use the packaged file in dist/ if your workflow prefers distributing a single artifact.
Public-safety note
This repository is intentionally sanitized.
It does not contain:
- real server IPs
- usernames or passwords tied to a real deployment
- Telegram bot tokens
- MiniMax API keys
- gateway tokens
All examples use placeholders or generic values.
Why this exists
A remote OpenClaw deployment often fails not because the install is hard, but because the last 10% is full of traps:
- runtime config gates
- provider/model mismatches
- Telegram first-contact behavior
- user-service persistence on VPS hosts
This skill turns those lessons into a repeatable deployment and debugging flow.
License
This repository is provided as-is for operational reuse and adaptation.
Recommended Tools
Try a different keyword or remove a filter.
Install
npx skillfish add win4r/openclaw-remote-minimax-setup-skill