Skills from clawhub.ai/thomaslwang

This skill scans installed OpenClaw skills by reading skill files from `~/.openclaw/skills/` and related directories and sending code chunks to an external model at `https://api.openguardrails.com/v1/model/chat/completions`. It runs shell commands like `ls`/`curl` and includes a hardcoded API key `sk-xxai-model-0e5a52bd1c70cca03d5f67fe1c2ca406` used in `Authorization` headers.

test

flaw0 is a CLI/library that scans OpenClaw code, plugins, skills, and Node.js dependencies and produces JSON or human-readable audit reports. The skill includes explicit shell instructions (e.g., `flaw0 scan`), requires the `ANTHROPIC_API_KEY` env var, and references external URLs such as `https://console.anthropic.com/` and GitHub.

This plugin detects hidden prompt-injection in long content for OpenClaw. It includes a hardcoded API key `sk-xxai-model-0e5a52bd1c70cca03d5f67fe1c2ca406`, calls `https://api.openguardrails.com/v1/model/`, and contains shell commands like `curl -L -o /tmp/test-email.txt https://raw.githubusercontent.com/openguardrails-for-openclaw/openguardrails-for-openclaw/main/samples/test-email.txt`.

MoltGuard is an OpenClaw plugin that analyzes long content for prompt injection using an external LLM. It embeds a hard-coded API key `sk-xxai-model-0e5a52bd1c70cca03d5f67fe1c2ca406`, calls `https://api.openguardrails.com/v1/model/`, and documents shell commands like `curl -L -o /tmp/test-email.txt`.