eastmoney_stock_simulator 安全扫描报告 — 可信 | ClawSafe

5 /100

eastmoney_stock_simulator

妙想提供的股票模拟组合管理系统，支持持仓查询、买卖操作、撤单、委托查询、历史成交查询和资金查询

A legitimate mock stock trading simulator skill that reads MX_APIKEY from environment, calls a declared dfcfs.com API for mock portfolio operations, and writes results to a declared output directory. No malicious behavior, obfuscation, credential harvesting beyond API usage, or undeclared capabilities detected.

技能名称eastmoney_stock_simulator

分析耗时26.5s

引擎pi

✓

可以安装

This skill is safe to use. The only minor concern is that the requests library version is unpinned in requirements.txt, which could theoretically allow a supply-chain downgrade. Consider pinning requests to a known-safe version for defense-in-depth.

安全发现 1 项

严重性	安全发现	位置
低危	Unpinned requests dependency 供应链 requirements.txt contains 'requests' without a version constraint. This allows pip to install any version, including those with known security vulnerabilities. `requests` → Pin requests to a specific version, e.g., 'requests>=2.32.0' to ensure known vulnerabilities are patched.	`scripts/requirements.txt:1`

资源类型	声明权限	推断权限	状态	证据
网络访问	`WRITE`	`WRITE`	✓ 一致	make_request() uses requests.post() to POST JSON to dfcfs.com — declared in SKIL…
文件系统	`WRITE`	`WRITE`	✓ 一致	save_result() writes .txt and .json files to /root/.openclaw/workspace/mx_data/o…
环境变量	`READ`	`READ`	✓ 一致	os.environ.get('MX_APIKEY') and os.environ.get('MX_API_URL') — both declared in …
命令执行	`NONE`	`NONE`	—	No subprocess, os.system, or shell command invocation found in scripts/mx_stock_…

2 项发现

🔗

中危外部 URL 外部 URL

https://mkapi2.dfcfs.com/finskillshub

SKILL.md:16

🔗

中危外部 URL 外部 URL

https://dl.dfcfs.com/m/itc4

SKILL.md:104

目录结构

3 文件 · 34.5 KB · 855 行

Markdown 1f · 471L Python 1f · 383L Text 1f · 1L

├─ ▾ 📁 scripts

│ ├─ 🐍 mx_stock_simulator.py Python 383L · 13.8 KB

│ └─ 📄 requirements.txt Text 1L · 9 B

└─ 📝 SKILL.md Markdown 471L · 20.7 KB

依赖分析 1 项

包名	版本	来源	已知漏洞	备注
`requests`	`*`	pip	否	Version not pinned — could resolve to a vulnerable version

安全亮点

✓ All capabilities (network, filesystem write, environment read) are declared in SKILL.md

✓ No shell execution, subprocess, os.system, or popen calls found

✓ No credential harvesting beyond the single API key needed for the declared service

✓ No base64-encoded payloads, eval(), or dynamic code execution

✓ No access to sensitive paths (~/.ssh, ~/.aws, .env, etc.)

✓ No data exfiltration or C2 communication — all network calls go to the declared dfcfs.com API

✓ File writes are confined to the documented output directory /root/.openclaw/workspace/mx_data/output/

✓ No obfuscation techniques detected

✓ No hidden HTML comments or injected instructions

✓ No persistence mechanisms (no cron, startup hooks, or backdoor installation)