CSV Wizard
PassAudited by ClawScan on May 1, 2026.
Overview
CSV Wizard appears to be a straightforward CSV-cleaning skill, with a packaging/provenance note because the actual CLI implementation is not included and the README points to an external GitHub repo.
This looks benign for normal CSV-cleaning use. Use a separate output file or backups for important datasets, and if you follow the README’s GitHub clone instructions, review the external code or pin a trusted version before running it.
Findings (2)
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.
A user could unintentionally produce a cleaned file that omits or changes data if they choose the wrong options.
The documented command can change CSV contents by removing rows and filling values, then write an output file. This is expected for a CSV-cleaning tool, but users should verify the output before relying on it.
/clean-csv sales.csv --drop-duplicates --fill-missing median --output clean-sales.csv
Run it on copies or specify a separate output file, and review the cleaned CSV before replacing important data.
If the external repository changes or is not trustworthy, running its CLI could expose local files or alter data outside what was reviewed here.
The README points users to external code that is not included in the submitted artifact set. This is user-directed rather than automatic, but it means users should review the external repository before running it.
git clone https://github.com/kimi-claw/skill-csv-wizard.git cd skill-csv-wizard ./bin/clean-csv --help
Inspect the external repository before use, prefer a trusted pinned commit or release, and avoid running unreviewed code on sensitive CSV files.