OpenClaw Setup on AWS (Free Tier) - Memory Upgrade

Set up a complete OpenClaw personal AI assistant from scratch using Claude Code. Walks through AWS provisioning, OpenClaw installation, Telegram bot creation, API configuration, Google Workspace integration, security hardening, and all power features. Give this to Claude Code and it handles the rest.

This skill will ask you for highly sensitive credentials (AWS account access, SSH private key, Anthropic/OpenAI/Groq keys, Telegram bot token, Google OAuth client_secret) and instruct an agent to use them to log into and configure your server. The metadata does not declare these secrets — a red flag. Before using it: (1) do NOT send your private .pem or long-lived master credentials to an AI; run SSH commands yourself or paste only the exact commands into your terminal; (2) prefer creating limited-scope, least-privilege API keys or IAM users (or temporary credentials) for setup; (3) restrict SSH access to your IP and use fail2ban/2FA; (4) verify the npm 'openclaw' package and any git repositories manually before installing; (5) consider performing Google OAuth client setup yourself and not handing client_secret.json to the agent; (6) if you are unsure, perform the install manually following the guide or ask a trusted human operator. The mismatch between declared metadata and the instructions is suspicious — proceed cautiously or avoid sharing secrets with the agent.

Type: OpenClaw Skill Name: openclaw-setup Version: 1.0.0 The skill instructs the AI agent to collect highly sensitive information (AWS account access, Anthropic, Groq, OpenAI API keys, Telegram bot token, Google Cloud credentials) directly from the user. It then directs the agent to SSH into the user's server using a provided `.pem` key and execute numerous `sudo` commands, including fetching and executing scripts from external sources (`curl -fsSL ... | sudo -E bash -` for Node.js, `git clone` and `make build` for `gogcli`). While these actions are plausibly necessary for the stated purpose of setting up a personal AI assistant on AWS, the extensive collection of credentials and the broad, privileged execution capabilities granted to the AI agent on the user's server represent significant security risks, classifying it as suspicious due to high-risk capabilities without clear malicious intent. All these instructions are found in `SKILL.md`.