扫描报告
10 /100
AI Cluster Pre-flight Check
Pre-flight check for GPU cluster nodes — 26 health checks covering GPU, PCIe, RDMA/IB, Docker, IOMMU, NUMA, firewall, and more
AI Cluster Pre-flight Check is a legitimate GPU cluster node health validation tool performing 26 read-only hardware/network diagnostics. No malicious behavior, credential harvesting, or covert data exfiltration found.
可以安装
No blocking action needed. The eval in check_1_25 should be documented explicitly, and dmidecode/iptables root requirements should be stated in SKILL.md for transparency.
安全发现 2 项
| 严重性 | 安全发现 | 位置 |
|---|---|---|
| 低危 | eval on user-supplied SWITCH_CLI_CMD not declared 文档欺骗 | lib/checks.sh:272 |
| 低危 | Root privilege requirements not declared 文档欺骗 | lib/checks.sh:213 |
| 资源类型 | 声明权限 | 推断权限 | 状态 | 证据 |
|---|---|---|---|---|
| 命令执行 | WRITE | WRITE | ✓ 一致 | SKILL.md declares 'bash' as required binary; preflight.sh uses bash to invoke GP… |
| 文件系统 | READ | READ | ✓ 一致 | Reads /proc/cmdline, /sys/class/net, /sys/class/infiniband, /sys/module; touch/r… |
| 网络访问 | READ | READ | ✓ 一致 | Cross-node SSH and ping checks declared in SKILL.md; cross-node section of check… |
| 环境变量 | NONE | READ | ✓ 一致 | Reads PREFLIGHT_*, MOUNT_POINT, SWITCH_* env vars — all declared in SKILL.md env… |
| 剪贴板 | NONE | NONE | — | Not used |
| 浏览器 | NONE | NONE | — | Not used |
| 数据库 | NONE | NONE | — | Not used |
| 技能调用 | NONE | NONE | — | Not used |
3 项发现
中危 外部 URL 外部 URL
https://clusterready.xperf.ai/ SKILL.md:19 中危 外部 URL 外部 URL
https://xperf.ai/ SKILL.md:24 提示 邮箱 邮箱地址
[email protected] SKILL.md:138 目录结构
5 文件 · 30.7 KB · 957 行 Shell 4f · 819L
Markdown 1f · 138L
├─
▾
lib
│ ├─
checks.sh
Shell
│ ├─
helpers.sh
Shell
│ └─
parser.sh
Shell
├─
preflight.sh
Shell
└─
SKILL.md
Markdown
安全亮点
✓ No credential harvesting — does not access ~/.ssh, ~/.aws, .env, or iterate os.environ for secrets
✓ No data exfiltration — all output is diagnostic; JSON results printed to stdout for the calling agent
✓ No obfuscation — all shell scripts are plain text, no base64, no eval-of-encoded strings
✓ Docker pulls from official nvidia/cuda and rocm/rocm-terminal images — known-good sources
✓ Cross-node capabilities (SSH, ping) are explicitly declared in the check catalog and documentation
✓ Skippable failure classification is a thoughtful UX feature that reduces false positives
✓ No remote script execution (no curl|bash, wget|sh patterns)
✓ No supply chain risks — no external dependencies beyond bash, jq, and system tools