social-proof-strength-audit
Skill2 starsChanged 4 months agoMITView on GitHub
Checks whether the proof on the page would convince a sceptic in the reader's own situation, or whether it is decoration. Use when a page has logos and testimonials and still feels unproven, or when proof was inherited from an older version of the business.
In this repository26 other files it gives its agents.
- AGENTS.mdAGENTS.md
- above-the-fold-clarity-reviewSkill
- accessibility-conversion-blocker-checkSkill
- comparison-page-positioning-reviewSkill
- conversion-leak-finderSkill
- cta-clarity-checkSkill
- form-friction-finderSkill
- google-ads-landing-page-experience-reviewSkill
- hero-section-diagnosisSkill
- landing-page-ab-test-readoutSkill
- landing-page-copy-readability-passSkill
- landing-page-scale-readiness-checkSkill
- landing-page-triageSkill
- lead-form-sales-handoff-checkSkill
- mobile-conversion-reviewSkill
- objection-map-builderSkill
- offer-clarity-diagnosisSkill
- page-length-fit-checkSkill
- page-speed-impact-reviewSkill
- paid-traffic-message-match-auditSkill
- popup-and-overlay-timing-reviewSkill
- pricing-page-clarity-reviewSkill
- thank-you-page-opportunity-auditSkill
- traffic-temperature-match-reviewSkill
- trial-vs-demo-path-decisionSkill
- trust-signal-auditSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
every proof element on the page: testimonials with whatever attribution they carry, logos, ratings, counts, case study links, …
- Distinctive: line 22
If a claimed number (customers, users, businesses served) has no definition or date, treat it as unverified and say so.
- Very specific: line 24
Not for: finding the objections in the first place (objection-map-builder), or reviewing security and risk-reversal signals …
- Distinctive: line 41
Do not recommend inventing, embellishing or composing a testimonial.
- Distinctive: line 64
Never draft, embellish or invent a testimonial, a rating or a customer number.
- Distinctive: line 65
Do not treat a logo wall as proof of results.
- Distinctive: line 66
Do not claim a lift from a proof change.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
Social Proof Strength Audit
Use this skill when
- testimonials, logos, ratings, case numbers or reviews are present on the page and the team is unsure any of them are doing work.
- the page carries logos and testimonials and still feels unproven.
- the proof was collected years ago, for a product or an audience that has since changed.
This skill judges the proof that exists. To find which objections need proof in the first place, use objection-map-builder.
Required input
- every proof element on the page: testimonials with whatever attribution they carry, logos, ratings, counts, case study links, certifications.
- where each one sits on the page relative to the offer and the ask.
- who the reader is, so proof relevance can be judged rather than proof volume.
- optional: which objection each proof element was meant to answer, if anyone decided that.
- If a claimed number (customers, users, businesses served) has no definition or date, treat it as unverified and say so.
Not for: finding the objections in the first place (objection-map-builder), or reviewing security and risk-reversal signals (trust-signal-audit).
Analysis workflow
- Inventory proof elements: logos, testimonials, case studies, metrics, reviews, screenshots, certifications and named customers.
- Classify proof by relevance to the ICP, specificity, recency, credibility and proximity to the CTA.
- Identify proof gaps for the main objection or risk on the page.
- Flag fake-looking, generic or misaligned proof that may reduce trust.
- Recommend where to add, move, rewrite or remove proof.
Decision rules
- Relevance beats volume. One testimonial from the reader's own segment outperforms eight logos from companies they cannot relate to, so rank by resemblance to the reader.
- Attribution is the whole of a testimonial's value. First name only is discounted to zero by a sceptical reader; name, role and company is a claim someone could check.
- A count with no definition and no date is marketing, not proof. Say which of the two is missing.
- Proof placed away from the objection it answers is not doing work. Report position, not just presence.
- Proof above the fold competes with the offer for the first three seconds. That is a trade-off worth naming, not automatically a defect.
- Do not recommend inventing, embellishing or composing a testimonial. If the proof does not exist, the recommendation is to go and get it, and from whom.
Output format
| Proof element | Attribution | Objection it answers | Position | Resembles the reader | Verdict |
| Quoted or named | Full / partial / none | The specific objection or none | Where on the page | Yes / no / unknown | Keep / re-attribute / move / replace / cut |
End with:
Strongest proof on the page:which one, and whyObjections with no proof against them:the listDecision:re-attribute / move / go and collect specific proof / ask for dataMissing data:definitions and dates behind any claimed number
Practical example
User: "We have 8 client logos above the fold, three testimonials with first names only, and a line saying Trusted by 500+ businesses."
Assistant should note: first-name-only testimonials carry no verifiable identity, so a sceptical reader discounts them entirely, and the fix is a full name, a role and a company on at least one of them; the 500+ number has no definition of what counts as a business or a date, so it reads as rounded marketing rather than a fact; eight logos above the fold compete with the headline for the first three seconds while telling the visitor nothing about whether those clients resemble them. Then: proof relevance beats proof volume, so the recommendation is one identified testimonial from the reader's own segment placed next to the objection it answers, and the logo wall moved below the offer. Whether that lifts CVR is a hypothesis; no per-section engagement data was supplied.
Guardrails
- Never draft, embellish or invent a testimonial, a rating or a customer number.
- Do not treat a logo wall as proof of results. It is proof of a relationship.
- Do not claim a lift from a proof change.
- Mark undefined and undated numbers as unverified rather than softening them.
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.

