openclaw-deploy

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly asks the agent to request API keys and Bot tokens and then embed those secrets verbatim into commands and configuration files (JSON apiKey, openclaw config set, curl requests with tokens), so the LLM would need to handle and output secret values directly.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). The set includes a high-risk direct-install URL (https://openclaw.ai/install.sh) intended to be piped to bash and raw GitHub-hosted scripts from a small/unknown user (raw.githubusercontent.com/ythx-101/...), which — despite many legitimate-looking API endpoints and localhost addresses — are typical vectors for malware distribution and thus make this a suspicious download source.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill includes runtime commands that fetch and execute remote code (e.g., curl -fsSL https://openclaw.ai/install.sh | bash and curl -sL https://raw.githubusercontent.com/ythx-101/x-tweet-fetcher/main/scripts/fetch_tweet.py -o ... plus the NodeSource setup scripts https://deb.nodesource.com/setup_22.x and https://rpm.nodesource.com/setup_22.x) which directly execute external code during installation, so they are high-confidence runtime-executed dependencies.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). 该技能指示在远程服务器上安装软件（通过包管理器）、运行 curl|bash 安装脚本、执行 openclaw gateway install 并写入/重载 systemd/user 配置、调用 loginctl/systemctl 等会改变系统服务和配置（并可能需要提升权限），因此会导致对主机状态的修改和潜在风险。