reviews
Skill5 starsChanged 37 days agoMITView on GitHub
Use when planning how to get product reviews or feedback, when to ask, what to do about the average rating, or how to respond to reviews and public complaints — e.g. "when should we send the review request", "should we remove this 1-star review", "how do we reply to this complaint", "we need more reviews", "our rating is 4.9".
1 other project uses a version of this.
Adopted by
Most recently taken up by
- dupe-com/skillstheir own version, first seen Oct 13, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- dupe-com/skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository18 other files it gives its agents.
- checkout-uxSkill
- digital-product-pricingSkill
- discountsSkill
- event-pricingSkill
- increase-aovSkill
- increase-conversionSkill
- increase-repeat-purchasesSkill
- post-purchaseSkill
- price-researchSkill
- pricingSkill
- product-copySkill
- product-pageSkill
- product-visualsSkill
- promotionsSkill
- raise-pricesSkill
- referralsSkill
- store-auditSkill
- subscription-pricingSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 32
Ask at 9–13 days: after purchase or delivery.
- Distinctive: line 33
Lead with a positive open-ended question: Before any rating: "What did you enjoy most about [product]?" (free text).
- Distinctive: line 37
Reply to every review: When companies started responding, ratings rose 0.12 stars and review volume 12%.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Reviews
Get more reviews and answer complaints in a way that wins over the people watching. Counterintuitive parts: an imperfect rating sells more than a perfect one, asking too soon reduces reviews, and the best first question is not "rate us".
When to use
- The review-request email or sequence
- A negative review or a too-perfect average
- Replying to a public complaint
- A product launch that needs a healthy review profile
When not to use
- Referral mechanics →
referrals - Loyalty, returns, packaging →
post-purchase
Inputs
Collect, or state your assumption:
- Product and how long it takes to judge (instant: soda; slow: skincare, a course)
- Current ask timing and wording; average rating; reviews the user wants to remove
- Whether reviews get replies; complaint volume, channel, and tone (polite vs rude)
Procedure
- Ask at 9–13 days after purchase or delivery. A request at 24 hours reduced reviews, worst among younger customers. Earlier only for products judged instantly; later for products that take weeks to evaluate.
- Lead with a positive open-ended question. Before any rating: "What did you enjoy most about [product]?" (free text). Responders spent 131% more afterward; even receiving the request helps. Usual questions follow. Automate it.
- Never gate the public review by score. Routing low scores to a private form and high scores to the review site is unsupported by the evidence and treated as deceptive by regulators (US FTC, UK CMA). Everyone gets the same path.
- Get reviews in early at launch. Ask all early customers as soon as step 1 allows — the same path for everyone, no picking the happy ones (that is selective solicitation, see step 3). The point is speed: a negative first review left the average 0.29 stars lower a year later on Amazon and Best Buy, so make sure the first review isn't left to chance by waiting.
- Leave the imperfection. 4.0–4.5 stars sells more than 4.5–5.0; perfect looks suspicious. A negative review irrelevant to most buyers helps: an espresso machine with four 5-stars rated 16% better with an irrelevant 1-star added. † Reply; don't remove.
- Reply to every review. When companies started responding, ratings rose 0.12 stars and review volume 12%.
- Public complaints: once, fast, warm, then private. You're replying to everyone reading, not the complainer. Reply fast and only once: empathy, an explanation if you have one, a private channel to continue. Replied-to posts get boosted on X and LinkedIn, so that one reply is what people see. High volume → a separate support account.
- Rude complaints: humor about the situation, never the person. Onlookers engaged more and were 18.3% more likely to buy than with no response. If it isn't funny, use step 7.
- Describe the product as an experience in copy (people sharing coffee, not bean origin) — such products collect more reviews. †
Output
- Solicitation plan: trigger, day offset, first question, follow-ups.
- Rating policy: what to leave, reply wording for an irrelevant negative.
- Templates: public complaint reply with private handoff; rude-complaint reply with situational humor; positive-review reply.
Common mistakes
- Asking 1 day after delivery — reduces reviews; 9–13 days.
- Opening with "rate 1–5" — open with what they enjoyed most.
- Score-gating — unsupported and legally risky.
- Removing an irrelevant 1-star — it makes the rest credible.
- Protecting a 4.9 — 4.0–4.5 sells more.
- Replying to a complaint repeatedly in public — once, then private.
Quick reference
| Situation | Do |
| Request timing | 9–13 days (earlier if judged instantly, later if slow) |
| First question | "What did you enjoy most?" |
| Launch | Ask all early customers promptly; same path for everyone |
| Average rating | 4.0–4.5; leave irrelevant negatives |
| All reviews | Reply to each |
| Public complaint | One fast empathetic reply → private |
| Rude complaint | Humor about the situation |
† No primary citation on file for this rule — a practitioner tip, not a measured finding.
Evidence and effect sizes: see reference.md.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

