clawschool 安全扫描报告 — 可疑 | ClawSafe

35 /100

clawschool

龙虾学校智力测试 — AI agent IQ benchmark that fetches questions from clawschool.teamolab.com, executes tasks, and submits results for scoring

Skill has an undocumented tool requirement (exec curl) and undeclared shell/filesystem write permissions, but the actual functionality is a benign AI benchmark with no malicious patterns.

技能名称clawschool

分析耗时32.1s

引擎pi

⚠

谨慎使用

Add explicit permission declarations for shell execution and filesystem write in the skill manifest. Verify the external domain clawschool.teamolab.com is trusted before usage.

安全发现 4 项

严重性	安全发现	位置
中危	Undeclared shell execution requirement 文档欺骗 SKILL.md explicitly mandates 'exec curl' for all HTTP requests but does not declare shell:WRITE permission in its manifest. This creates a doc-to-code mismatch where the skill's actual execution path requires shell access that is not declared. 所有 HTTP 请求必须使用 `exec curl`，禁止使用 web_fetch、web_search、browser 等工具 → Add a permissions section or allowed-tools declaration to SKILL.md header specifying shell:WRITE for network requests.	`SKILL.md:15`
中危	Undeclared filesystem write operations 文档欺骗 Skill creates a directory at ~/Desktop/claw_evidence and writes a temp JSON file to /tmp/clawschool_submit.json. Neither operation is declared in the skill's permission manifest. `cat > /tmp/clawschool_submit.json << 'EOFSUBMIT'` → Declare filesystem:WRITE in the skill manifest for the ~/Desktop/claw_evidence and /tmp paths.	`SKILL.md:63`
低危	Unresolved variable placeholders 文档欺骗 {{TOKEN}} and {{LOBSTER_NAME}} placeholders are used throughout but never explained how they are populated. This raises questions about where credentials originate and whether they are handled securely. `{{TOKEN}}, {{LOBSTER_NAME}}` → Document how tokens and lobster names are obtained, stored, and whether they are persisted or shared externally.	`SKILL.md:28`
低危	Access to user home directory path 权限提升 The skill writes to ~/Desktop/claw_evidence, which is a user-owned path. While legitimate, this is an undeclared filesystem access to a path outside of the typical sandboxed workspace. `mkdir -p ~/Desktop/claw_evidence` → Consider scoping file operations to a sandboxed directory and declare the path in the manifest.	`SKILL.md:38`

资源类型	声明权限	推断权限	状态	证据
命令执行	`NONE`	`WRITE`	✗ 越权	SKILL.md: '所有 HTTP 请求必须使用 exec curl'
文件系统	`NONE`	`WRITE`	✗ 越权	SKILL.md: 'mkdir -p ~/Desktop/claw_evidence' and 'cat > /tmp/clawschool_submit.j…
网络访问	`NONE`	`WRITE`	✗ 越权	SKILL.md: Makes POST requests to external API without declared network:WRITE

2 项发现

🔗

中危外部 URL 外部 URL

https://clawschool.teamolab.com/api/test/start?token=

SKILL.md:43

🔗

中危外部 URL 外部 URL

https://clawschool.teamolab.com/api/test/submit

SKILL.md:79

目录结构

1 文件 · 4.6 KB · 127 行

Markdown 1f · 127L

└─ 📝 SKILL.md Markdown 127L · 4.6 KB

安全亮点

✓ No base64-encoded commands or obfuscation observed

✓ No credential harvesting loops (no iteration over os.environ for sensitive keys)

✓ No curl|bash or wget|sh remote script execution patterns

✓ No access to ~/.ssh, ~/.aws, or .env sensitive paths

✓ No reverse shell, C2 communication, or data exfiltration patterns

✓ TLS verification is explicitly enforced (no -k or --insecure flags)

✓ No hidden HTML comments or prompt injection payloads detected

✓ Skill behavior is consistent with a legitimate AI benchmarking tool