名称: openclaw-security-audit
描述: 使用用户配置的 LLM 审计 OpenClaw 配置的安全风险并生成修复报告。
元数据:
openclaw:
requires:
bins: ["cat", "jq"]
os: ["darwin", "linux", "windows"]
这是一个纯本地技能,用于审计 ~/.openclaw/openclaw.json 配置文件,执行 15 项以上的安全检查,并利用用户现有的 LLM 配置生成详细报告。无需外部 API 或密钥。
cat, jq)读取配置文件。~/.openclaw/openclaw.jsongateway.bind 设置(例如 0.0.0.0 未配置适当认证)allowFrom){
"config_path": "~/.openclaw/openclaw.json",
"openclaw_version": "present",
"gateway": {
"bind": "0.0.0.0",
"auth_token": "missing"
},
"channels": {
"allowFrom": "missing",
"rate_limits": "missing"
},
"secrets": {
"hardcoded": "detected"
},
"tool_policies": {
"elevated": "unrestricted"
}
}
报告必须包含:
read_config_path = input.target_config_path || ~/.openclaw/openclaw.json
raw_config = cat(read_config_path)
json = jq parse raw_config
metadata = extract_security_metadata(json)
findings = build_findings(metadata)
report = openclaw.agent.analyze(findings, format=markdown)
return report