Skills from clawhub.ai/phenixstar

Dangerous skill for OpenClaw diagnostics and auto-fix that runs CLI checks, installs packages, and applies fix recipes. It executes shell commands (e.g., `pip install`, `wget`, recipe `command` steps), performs network calls to `https://docs.openclaw.ai` and `api.openai.com`, and reads/stores secrets like `OPENAI_API_KEY` into `~/.openclaw/openclaw.json`.

This skill provides unified discovery, security vetting, syncing, and installation tooling for OpenClaw (search, vet, sync, status, install). It invokes local CLIs (`npx`, `gh`, `python3`) and performs network requests (e.g., `https://raw.githubusercontent.com/VoltAgent/awesome-openclaw-skills/master/README.md`) while scanning files under `Path.home() / ".openclaw" / "workspace"`.

High-risk skill: its code executes local shell commands via `subprocess.run`/`run_command` and runs fix recipes that can invoke arbitrary commands, while also performing network calls to `https://docs.openclaw.ai` and `api.openai.com`. It prompts for and stores API keys into `openclaw.json`/`~/.openclaw` and references env vars like `OPENAI_API_KEY`.