🦞 OpenClaw Starter Guide

OpenClaw 小白养成手册。从零开始搭建多 Agent AI 助手系统的完整指南，包含免费起步方案（Qwen 零成本）、进阶方案（MiniMax Coding Plan ¥49/月）、旗舰外援（SiliconFlow/NewCLI 按需调用）。涵盖模型策略、Fallback 链设计、额度管理、常见故障排查。适合首次部署 OpenClaw 或想优化模型成本的用户。当用户说"怎么开始"、"新手指南"、"省钱方案"、"模型怎么选"、"fallback 怎么配"时使用此 skill。

This skill is an instructional guide and appears coherent and low-risk. Before following steps: 1) Verify third-party provider links (SiliconFlow, MiniMax, NewCLI) and their terms — some links in README include affiliate params. 2) Do not paste real API keys into public logs or chat; store keys in your OpenClaw/provider config and follow least-privilege practices. 3) Treat claims like “free/unlimited calls” or exact quota numbers as marketing — validate by testing in a controlled environment. 4) When installing clawhub or provider packages, review the packages' sources (npm/package homepage) to ensure they are legitimate. If you want higher assurance, provide the skill author/source or official vendor documentation so we can re-evaluate if anything changes.

Type: OpenClaw Skill Name: openclaw-starter-guide Version: 1.2.0 The skill bundle is a comprehensive guide for setting up and managing OpenClaw AI agents. It provides instructions for users on model selection, fallback chains, cost management, and troubleshooting, including commands for installing the `clawhub` CLI and other OpenClaw provider skills. The `SKILL.md` contains diagnostic `curl` commands and system management commands, but these are clearly presented as instructions for the user to execute in their terminal, not for the AI agent to execute. There is no evidence of data exfiltration, malicious execution, persistence mechanisms, prompt injection against the agent, or obfuscation. The referral links to SiliconFlow and NewCLI are for legitimate services and are part of the guide's purpose to help users set up providers.