AI Data Analysis
PassAudited by ClawScan on May 10, 2026.
Overview
This appears to be a benign local CSV analysis tool, with low-risk notes about saved output copies, an undeclared pandas dependency, and documentation that overstates the implemented features.
This skill is reasonable to use for local CSV analysis if you trust the source and have the required Python packages. Be aware that outputs may retain sensitive data locally, and the provided implementation appears more limited than the advertised Excel, charting, and export features.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If you analyze sensitive data, a report or cleaned copy may remain on disk after the run.
The skill persists generated reports and cleaned data locally. This is expected for data analysis, but those files may contain summaries or copies of sensitive dataset contents.
OUTPUT_DIR = SCRIPT_DIR / "output" ... output_file.write_text(output, encoding='utf-8') ... df_clean.to_csv(output_file, index=False)
Run it only on intended files and review or delete generated files in the output directory when handling sensitive data.
The skill may fail unless pandas is already installed, or a user/agent may need to install it separately.
The code requires pandas, but the supplied artifacts include no install spec and package.json does not declare Python dependencies. This is an undeclared runtime dependency, not evidence of hidden installation.
import pandas as pd
Confirm the required Python packages and install them from trusted sources before use.
Users may expect capabilities that the provided code does not actually perform.
The documentation claims Excel, charting, and PDF/HTML export support, while the supplied main.py only implements CSV reading, summary statistics, and CSV/JSON cleaning output.
✅ 支持 CSV, Excel, JSON ... ✅ 自动生成图表（图表库） ... ✅ 可导出 PDF/HTML
Treat the skill as a basic CSV analysis/cleaning helper unless the missing features are implemented and reviewed.