agentleFS
MuvonMuvon/octomind-tap/skills/marketing-review-program/SKILL.md

marketing-review-program

Skill6 starsChanged 3 days agoApache-2.0View on GitHub

Plans how to ask real customers for honest public reviews: site choice, invitation copy, incentive and disclosure rules. Use to start a review program.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    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:

    Ask after the activation milestone or another measured success moment, never at signup.

    line 32
  2. Distinctive:

    If activation is not defined, the first deliverable is a definition: the events and time window that mark a customer who got value.

    line 33
  3. Distinctive:

    Start from where this buyer compares products.

    line 37
  4. Distinctive:

    When captured AI answers exist, note which review sites they cite.

    line 38
  5. Distinctive:

    No sentiment gating: don't send satisfied customers to the public site and unhappy ones to a private form.

    line 50
  6. Distinctive:

    Read on 2026-11-09; recheck each page before launch.

    line 55
  7. Distinctive:

    Not legal advice. Read at eCFR and the FTC Q&A.

    line 71
  8. Distinctive:

    Don't suppress, threaten or offer payment to remove.

    line 90

≈ “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

Overview

Reviews and customer posts are claims from third parties, so their value depends on being honest and verifiably unpaid for sentiment. This skill produces a program plan: when to ask, where, whom, with what wording, what incentives each platform allows, and how to measure it. It sends no invitations and changes no profile. Platform and legal positions below were read on 2026-11-09 and change; recheck before launch.

Mental model

Moment of success, then neutral invitation to every eligible customer, then honest review on a chosen site, then disclosed and permitted reuse. Each step has one failure: asking too early, asking selectively, rewarding sentiment, hiding the connection. The plan removes those four.

Rules

Sequence

  • Ask after the activation milestone or another measured success moment, never at signup. A review from someone who has not yet used the product is not a review of the experience.
  • If activation is not defined, the first deliverable is a definition: the events and time window that mark a customer who got value. Don't invite until it exists.

Choose sites

  • Start from where this buyer compares products. Ask the team, check sales calls and search results for the category.
  • When captured AI answers exist, note which review sites they cite. Label this an observation, not a ranking cause.
  • Pick one to three sites. Spread thin reviews across many sites help none.

Eligibility and frequency

  • Invite real customers with actual use. Exclude employees, relatives, investors and agents, or require clear disclosure of the relationship.
  • Cap invitations per person, for example one per product per 12 months. This is a convention, not a platform rule.
  • Contact basis follows the outreach-compliance skill: record the source, honor opt-outs, respect the recipient's jurisdiction.

Invitation

  • Neutral wording that asks for an honest review, states any incentive, and goes to every eligible customer the same way.
  • No sentiment gating: don't send satisfied customers to the public site and unhappy ones to a private form. Trustpilot and G2 name selective solicitation as a violation; the FTC says it can violate the FTC Act.
  • A private feedback channel may exist, offered to everyone alongside the public link.

Incentives by destination

Read on 2026-11-09; recheck each page before launch. Conditional means allowed under limits. The Capterra and Amazon rows come from search extracts of the official pages, because direct fetches failed.

DestinationStatusSource and rule
G2ConditionalGuidelines: incentives may not depend on sentiment; vendors may not incentivize independently without disclosing to G2; G2 labels incentivized reviews
Capterra / Gartner Digital MarketsConditionalGuidelines: nominal incentives only, offered regardless of rating and never contingent on a positive review; insiders and direct competitors can't receive them; use the Vendor Portal link; incentivized reviews are labelled
TrustpilotForbiddenGuidelines for businesses: no incentives for writing or editing reviews; invite everyone the same way; no selective invitations
Google Business ProfileForbiddenPolicy: no incentives; no discouraging negative reviews or selectively soliciting positive ones
Apple App StoreForbiddenGuidelines: paid or incentivized feedback is manipulation; apps must not force a rating or review for access
Google PlayForbiddenPolicy: do not offer rewards for ratings or reviews
AmazonForbiddenReview policy summary: selling partners can't offer compensation for a positive or negative review, or ask customers to edit or remove one. Amazon Vine is run by Amazon and isn't a seller incentive

Where a platform allows incentives, disclose them in the invitation and keep the offer independent of the rating.

Legal floor

Not legal advice. Read at eCFR and the FTC Q&A.

  • 16 CFR 465.4: no paying or rewarding reviews where the reward depends on a particular sentiment, stated or implied.
  • 16 CFR 465.5: reviews by officers, managers, employees, agents or their immediate relatives need clear disclosure; managers must not solicit them without it. The FTC says generalized solicitations to all customers are exempt.
  • 16 CFR 465.7: no unfounded legal threats, intimidation or knowingly false accusations to block or remove a review; no claiming displayed reviews are all reviews when some are hidden by rating.
  • 16 CFR 255.5: a material connection that the audience would not expect, including free product, discounts or other benefits, must be disclosed clearly. Incentives are not banned by the rule.
  • UK: Digital Markets, Competition and Consumers Act 2024, Schedule 20 paragraph 13 covers fake reviews, concealed incentives and selective publication; in force 6 April 2025.
  • EU: not verified here; flag for local review.

Recommend legal review before launch.

Social posts

  • Anyone paid, rewarded or given product to post must disclose the connection clearly in the post itself, in plain words, not only on a profile (16 CFR 255.5).
  • Use the platform label where one exists. TikTok: commercial content disclosure setting, shown as Paid partnership (help). LinkedIn: brand partnership label for content shared in exchange for value (help). Instagram: paid partnership label for content made for an exchange of value, including free products and gifts (help). X: Paid Partnership setting for compensated, gifted, or affiliate promotions (policy, reviewed 2026-09).
  • Get written permission before reusing a customer's post, quote or name. A reposting brand makes its own disclosures. For creators, see outreach-influencer.

Negative reviews

Don't suppress, threaten or offer payment to remove. Route the problem to support, then let the review stand. Writing public replies is owned elsewhere.

Measurement

Report counts with denominators: eligible customers, invitations sent, reviews per site, rating distribution for incentivized versus non-incentivized reviews, review recency, and how many reviewers reached activation. Rating drift upward after an incentive starts is a warning sign. Treat AI-answer changes as a separate observation with no causal claim.

Examples

Wrong invitation: "Love our product? Leave us 5 stars on Trustpilot for a $20 gift card!" Right: "You have used Acme for 30 days. If you are willing, tell others what it was like, good or bad: [link]. We do not reward reviews."

Wrong routing: score 9-10 goes to the public site, score 0-8 goes to a private form. Right: everyone receives the same message with the public link and a separate support link.

Wrong post brief: "Post a glowing video and get a free year." Right: "Free year offered for any honest post. Turn on the paid partnership label and say you received it."

Checklist

  • Activation trigger defined before any invitation timing?
  • One to three sites chosen with the buyer-comparison evidence?
  • Insiders excluded or disclosed; frequency cap set and labelled a convention?
  • One invitation per site, neutral, same for every eligible customer, incentive stated?
  • Every incentive checked against the table and the platform's live page?
  • Social posts carry in-post disclosure and the platform label where verified?
  • Written permission step for reuse?
  • Negative-review routing avoids suppression?
  • Metrics have denominators; legal review items listed?

Composition / References

Save as review-program-[product].md with: activation trigger, sites and policy check, eligibility, invitation copy per site, incentive and disclosure, social program, negative-review routing, metrics, legal review items.

Sources accessed 2026-11-09. The frequency cap and one-to-three site limit are practitioner conventions.

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.