Return Rate Reducer

PassAudited by ClawScan on May 1, 2026.

Overview

This skill appears to be a coherent return-rate analysis helper with no evidence of credential use, network sharing, persistence, or destructive behavior.

This skill looks safe to install for return-rate analysis. Before using it, treat returns exports and support tickets as potentially sensitive, redact unnecessary personal data, and run the optional scripts only on files and output locations you intentionally choose.

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.

What this means

Return exports or support-ticket data shared with the agent may include sensitive business or customer information.

Why it was flagged

The skill is designed to analyze user-provided return data; this is expected for the purpose, but those exports may contain business metrics or customer/order details.

Skill content
7. **Data access** — Can you share a CSV/export, or are you working from memory and screenshots?
Recommendation

Share only the fields needed for analysis and redact names, emails, addresses, payment details, or other unnecessary personal data.

What this means

If run, the script can read the chosen returns file and create or overwrite the chosen report file.

Why it was flagged

The helper script reads a user-selected local CSV and can write a generated report to a user-selected path; this is central to the skill and no network, shell, or hidden execution behavior is shown.

Skill content
parser.add_argument("--in", dest="in_path", required=True, help="Path to returns CSV.") ... parser.add_argument("--out", dest="out_path", default=None, help="Output file (default: stdout).") ... out_path.write_text(report, encoding="utf-8")
Recommendation

Run the helper only on intended files, review the output path before using `--out`, and avoid running it with unnecessary elevated permissions.