agentleFS
SkillMedevSkillMedev/ecommerce-dtc-ops/skills/review-to-faq-builder/SKILL.md

review-to-faq-builder

Skill1 starsChanged 40 days agoMITView on GitHub

Mines customer reviews and Q&A exports into a pre-purchase PDP FAQ and objection-handling block that answers shopper hesitation before it forms, ordered by conversion impact. Use when someone asks "turn these reviews into an FAQ", "shoppers keep asking the same questions before buying", "how do I address the negative reviews on the product page", or has a review or Q&A export and wants to write or rebuild a product page FAQ or cut returns from preventable surprises. Do NOT use to analyze support tickets or CSAT/NPS verbatims for quality root causes - use csat-root-cause instead; do NOT use to write the main product description - use product-description-writer instead.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 9 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    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
  1. Distinctive:

    If the review set is small (as a working floor, under ~30 reviews) or skewed (one channel, one cohort, incentivized), say so and recommend …

    line 23
  2. Distinctive:

    Read every one- and two-star review for the named deal-breaker and answer it directly.

    line 31
  3. Distinctive:

    Do not write an answer that contradicts what the reviews actually report - shoppers read both, and the mismatch destroys trust in the whole …

    line 73
  4. Distinctive:

    Do not invent testimonials, quotes, or buyer attributions.

    line 74
  5. Distinctive:

    Do not hide deal-breaker objections - an unanswered objection blocks the sale silently.

    line 75
  6. Distinctive:

    Do not paper over a real product defect with clever copy.

    line 76
  7. Distinctive:

    Do not publish FAQ claims from a tiny or skewed review set without flagging the limitation.

    line 77
  8. Distinctive:

    Do not order the FAQ by internal convenience (logistics first).

    line 78
  9. Distinctive:

    Every question maps to a doubt present in the review set, with its frequency count recorded.

    line 82

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Review-to-FAQ Builder

Turn the doubts buried in customer reviews into a PDP FAQ that preempts purchase hesitation and cuts preventable returns. The costly mistake this prevents is the marketing-written FAQ - questions nobody asks, answered in brand voice - while the real deal-breakers ("does it run small?", "does the smell fade?") sit unanswered in the one-star reviews, blocking sales and generating returns.

Operating procedure

Order matters: clustering before writing prevents cherry-picking, and frequency ranking is what separates a real FAQ from a wishlist.

Step 1: Gather inputs

  • The review and Q&A export (all star levels - the one- and two-star reviews are the most valuable input, not an embarrassment to skip).
  • The current return and shipping policy, verbatim, so answers never contradict it.
  • Any existing FAQ, to preserve what already works.
  • Product facts: dimensions, materials, compatibility - for verifying answers.

If the review set is small (as a working floor, under ~30 reviews) or skewed (one channel, one cohort, incentivized), say so and recommend collecting more before publishing claims from it.

Step 2: Cluster the recurring doubts

Group raw reviews into themes - sizing/fit, durability over time, setup difficulty, compatibility, value vs price, smell/taste/material, shipping speed, return ease - and count frequency per cluster. The top 5-8 clusters by volume become the FAQ. A doubt that appears once is noise; a doubt in 10% or more of reviews is mandatory.

Step 3: Convert negative reviews into honest answers

Read every one- and two-star review for the named deal-breaker and answer it directly. If reviewers say "runs small," the answer is "Yes - many customers size up; see our fit guide," not silence or spin.

Step 4: Write questions in the customer's words

Phrase each question the way a shopper types it, using actual review vocabulary ("Will this fit a queen mattress?" not "Dimensional compatibility"). This also matches voice search and "People also ask" queries.

Step 5: Write tight answers

2-4 sentences each: lead with the direct yes/no/number, then the supporting detail. An answer that opens with brand throat-clearing gets skimmed past at the exact moment of hesitation.

Step 6: Quote real buyers as proof

Where a review answers a question better than you can, quote it - attributed as a verified buyer, lightly edited for clarity only. Use only real quotes; a fabricated testimonial is both a conversion risk and, in many jurisdictions, a legal one.

Step 7: Order by conversion impact

Lead with the objection that most often blocks the sale - fit, function, "will it work for me" - and put logistics (shipping, returns) lower. End with one reassurance line tied to the actual policy.

Worked artifact: bad/good transformation pair

Source review (2 stars): "Looks great but the fabric is way thinner than the photos suggest. Almost see-through in daylight. Returned it."

Bad FAQ entry:

Q: What is the fabric like? A: Our curtains are crafted from premium lightweight polyester voile, designed with an airy, elegant drape that elevates any living space.

Why it fails: the question is not what shoppers ask, the answer dodges the actual doubt (sheerness), and the buyer discovers the truth after purchase - as a return.

Good FAQ entry:

Q: Are these curtains see-through? A: Partially - they're a semi-sheer voile that filters light rather than blocking it. In direct daylight you'll see silhouettes but not detail. If you need privacy or darkness, pair them with our blackout liner. As one verified buyer put it: "Perfect for the living room, but I added the liner in the bedroom."

Why it works: the question uses review vocabulary, the answer leads with the direct truth, sets accurate expectations (fewer returns), routes to a legitimate upsell, and borrows a real buyer's credibility.

Deliverable

Produce a PDP-ready FAQ block containing: 5-8 questions in shopper vocabulary ordered by conversion impact, each with a 2-4 sentence direct-first answer, at least one real attributed buyer quote, one policy-tied reassurance line - plus a frequency table showing which review cluster each question came from and its count.

Do NOT

  • Do not write an answer that contradicts what the reviews actually report - shoppers read both, and the mismatch destroys trust in the whole page.
  • Do not invent testimonials, quotes, or buyer attributions.
  • Do not hide deal-breaker objections - an unanswered objection blocks the sale silently; an answered one converts the honest-fit buyer and deflects the return.
  • Do not paper over a real product defect with clever copy; flag it for the product team instead.
  • Do not publish FAQ claims from a tiny or skewed review set without flagging the limitation.
  • Do not order the FAQ by internal convenience (logistics first); order by what blocks purchase.

Quality bar

  • Every question maps to a doubt present in the review set, with its frequency count recorded.
  • Each answer is consistent with what reviews actually report and with the stated return/shipping policy.
  • Questions use shopper vocabulary; answers lead with the direct yes/no/number.
  • Hesitation-blocking objections appear before compliments and logistics.
  • At least one deal-breaker from the one- and two-star reviews is answered honestly.

Neighbors

Support-ticket and CSAT root-cause analysis: csat-root-cause. The product description itself: product-description-writer. Structuring answers for answer engines and featured snippets: aeo-answer-blockifier. Per-variant copy questions ("which size should I get") often route into variant-copy-scaler territory.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.