Review Analyzer

Extract sentiment patterns, repeated pain points, and feature requests from customer reviews to prioritize product fixes and copy improvements.

This skill appears coherent and focused, but consider these practical points before installing and using it: - Privacy: Do not paste reviews that contain personal data, emails, phone numbers, or other PII. The skill asks you to provide raw review text; sanitize sensitive fields first. - Sample size: The doc inconsistently mentions "20 or more" in one place and a minimum of "ten" in another; for reliable frequency-based patterns use 30+ reviews where possible. - Verified vs unverified: The skill says it won't distinguish verified purchases unless you include that info — include a flag column if that distinction matters. - Hallucination risk: Estimated implementation effort and predicted review-score impact are model-generated estimates; validate recommendations with real testing before making supplier or product decisions. - Auditability: Because this is instruction-only (no install), review outputs carefully and keep an audit trail of inputs/outputs if you need to justify changes to product or listings. - Autonomy: The skill is user-invocable and not forced always-on. If you are worried about autonomous runs, ensure platform-level controls for skill invocation are in place. Overall, nothing in the skill requests unrelated access or installs, so it is coherent with its stated purpose.

Type: OpenClaw Skill Name: review-analyzer Version: 1.0.0 The skill bundle consists of a metadata file and a markdown-based instruction set (SKILL.md) designed to guide an AI agent in analyzing customer reviews. There is no executable code, no external network calls, and no evidence of prompt injection or malicious intent. The instructions are strictly aligned with the stated purpose of sentiment analysis and product feedback extraction.