Return Rate Reducer
v0.1.0Reduce e-commerce return rates through data-driven root-cause analysis, product-page fixes, and policy optimization. Use this skill whenever the user mention...
⭐ 0· 207·0 current·0 all-time
byRIJOY-AI@rijoyai
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (reduce return rates) matches the included artifacts: a detailed SKILL.md, a PDP lint script, a return CSV analyzer, playbook and example files. Nothing in the manifest asks for unrelated services, binaries, or credentials.
Instruction Scope
Runtime instructions ask for standard e-commerce context (platform, return CSV, PDP content, return reasons) and limit scope to prevention-focused analysis. They do not instruct reading unrelated system files or pulling secrets. The included scripts operate on user-supplied files (markdown/CSV), which is coherent with the described workflow.
Install Mechanism
No install spec is provided (instruction-only plus lightweight Python scripts). There are no downloads, external installers, or archive extracts. The scripts are simple, local utilities (no package installation required).
Credentials
The skill declares no required environment variables, no primary credential, and no config paths. The code reads only user-supplied files; it does not reference environment secrets or remote endpoints.
Persistence & Privilege
Flags show always:false and normal autonomous invocation allowed. The skill does not request permanent system presence, nor does it modify other skills or system-wide settings.
Assessment
This skill appears coherent and safe to review, but follow these precautions before running or uploading data: 1) Sanitize uploads—remove customer PII (names, emails, addresses, payment info) from CSVs or product exports before sharing. 2) Review the included Python scripts locally first to confirm behavior (they only read CSV/markdown and produce reports). 3) Run scripts in an isolated environment if you are unsure (they have no network calls, but verify before executing). 4) Test on sample/anonymized data to validate outputs and thresholds before using real production exports. 5) If you plan to let an agent run autonomously with access to files, limit the data scope and revoke access after testing. If you want, I can point out exactly which CSV columns to strip or help produce an anonymized sample.Like a lobster shell, security has layers — review code before you run it.
latestvk975gm6197d54xhjmnxst86szd82q6m4
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
