策略分析报告师
v1.0.0量化策略评估报告生成器。读取 Excel 净值数据（策略净值 + 可选品种净值）， 自动计算 20+ 量化指标、生成 12+ 张专业图表，AI 撰写深度分析文字， 最终输出机构级 PDF（16+ 页）和 Word 策略评估报告。 触发：用户上传 Excel 净值数据并要求生成策略评估报告、分析报告。
⭐ 0· 91·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (quant strategy report generator) matches the included scripts and instructions: data loader, analyzer, charting, and PDF/DOCX report builders. No unrelated credentials, binaries, or config paths are requested.
Instruction Scope
SKILL.md confines runtime actions to reading user-provided Excel files under data/, running the analysis and report scripts, and writing outputs to output/. The AI is instructed to produce a structured content.json for the report; there are no instructions to access unrelated system files, env vars, network endpoints, or to exfiltrate data.
Install Mechanism
No install spec is provided (instruction-only metadata), and the SKILL.md lists reasonable Python package dependencies (pandas, numpy, matplotlib, reportlab, python-docx, etc.). No downloads from unknown URLs or extract/install operations are present.
Credentials
The skill requires no environment variables or credentials. It only uses local filesystem paths (data/ → output/). The declared dependencies are proportional to tasks (data processing, plotting, document generation).
Persistence & Privilege
always is false and the skill does not request persistent platform privileges. It does not modify other skills or system-wide configs in the code reviewed.
Assessment
This skill appears to do what it claims: load Excel NAV files, compute metrics, generate charts, and produce PDF/DOCX reports. Before use: (1) Run it in an isolated environment or virtualenv and install the listed Python packages; (2) review the Excel files you upload—they can contain sensitive financial data; (3) spot-check the generated output/content.json to ensure AI-written analyses don’t include undesired or sensitive text; (4) ensure reportlab >= 3.6 if you need proper Chinese fonts (SKILL.md notes STSong-Light fallback); (5) if you plan to run this on production or confidential data, review the full source locally (the included code is straightforward) and avoid running in an environment where arbitrary code execution would risk exposing other secrets or services.Like a lobster shell, security has layers — review code before you run it.
latestvk976chmnbeeegs7jnt7jeehva583e3re
License
MIT-0
Free to use, modify, and redistribute. No attribution required.