扫描报告
5 /100
A股短线交易决策 A-Share Short-Term Decision
A-share short-term trading decision skill for 1-5 day horizon using AkShare for real/historical market data
A legitimate A-share short-term trading analysis skill with clean code, no shell execution, no credential theft, and no data exfiltration. All capabilities align with documented behavior.
可以安装
This skill is safe to use. No security concerns identified.
| 资源类型 | 声明权限 | 推断权限 | 状态 | 证据 |
|---|---|---|---|---|
| 文件系统 | WRITE | WRITE | ✓ 一致 | SKILL.md declares data/ decision_log.jsonl writes; decision_eval.py:44-45 _appen… |
| 网络访问 | READ | READ | ✓ 一致 | AkShare data fetching - stock_zt_pool_em, stock_zh_a_hist, etc. in market_data.p… |
| 命令执行 | NONE | NONE | — | No subprocess or shell execution found |
| 环境变量 | NONE | READ | ✓ 一致 | debug_utils.py:16 reads SHORT_DECISION_DEBUG env; settings.py:40 reads SHORT_DEC… |
| 技能调用 | NONE | NONE | — | No skill invocation capabilities used |
| 剪贴板 | NONE | NONE | — | No clipboard access found |
| 浏览器 | NONE | NONE | — | No browser automation found |
| 数据库 | NONE | NONE | — | No database access; only JSONL file logging |
目录结构
18 文件 · 68.5 KB · 2111 行 Python 12f · 1572L
Markdown 3f · 448L
JSON 2f · 74L
YAML 1f · 17L
├─
▾
prompts
│ └─
__init__.py
Python
├─
▾
tools
│ ├─
__init__.py
Python
│ ├─
debug_utils.py
Python
│ ├─
decision_eval.py
Python
│ ├─
fusion_engine.py
Python
│ ├─
indicators.py
Python
│ ├─
market_data.py
Python
│ ├─
money_flow.py
Python
│ ├─
reporting.py
Python
│ ├─
risk_control.py
Python
│ └─
settings.py
Python
├─
_meta.json
JSON
├─
config.json
⚠
JSON
├─
main.py
Python
├─
README.md
Markdown
├─
requirements.md
Markdown
├─
scheduler.yaml
YAML
└─
SKILL.md
Markdown
依赖分析 2 项
| 包名 | 版本 | 来源 | 已知漏洞 | 备注 |
|---|---|---|---|---|
akshare | * | pip | 否 | Chinese financial data library; version not pinned |
pandas | * | pip | 否 | Standard data analysis library; version not pinned |
安全亮点
✓ Clean Python codebase with no shell execution capabilities
✓ No credential theft - does not access ~/.ssh, .env, or similar sensitive paths
✓ No data exfiltration - all network calls are read-only via AkShare for financial data
✓ No obfuscation or encoded payloads detected
✓ All filesystem writes are declared and necessary (data/decision_log.jsonl for prediction logging)
✓ Dependencies (akshare, pandas) are standard financial data analysis libraries
✓ Environment variable access is limited to debug flags only
✓ Code is well-structured with proper error handling