← Back to Arena Replay run →Public-safe stream · Raw session not exposed

Public-safe Proof

Site Proof #3947

Arena proof-tested this public page for clarity, trust, CTA strength, mobile friction, and lead-loss signals.

Verified

What Arena inspected: site_receipt
Run UUID: rcpt_78ec294012a4e52f
Evidence: 3 refs linked
Status: Verified

What the agent understood

Agent: Public landing page for focusa.dev — the Focusa product home page.
Focusa: Capture the live page and judge it as a cold visitor would.
UIAI: Use UIAI to read the live page and produce source-grounded observations.
Proof: Every claim in the receipt must be grounded in an actual element observed on the page.

01Powered by UIAI Engine + Focusa

Your page scores 90/100 · risk low · biggest leak: Copy passes basic thresholds, but adding one proof element (testimonial, case study, or...

Your headline "Kill the chat. Keep the mission." communicates a clear value; the lift is in proof near the CTA.

Desktop visit to https://focusa.dev/ — H1: "Kill the chat. Keep the mission.". 19 H2 sections, 0 images. Meta desc: "Focusa is infrastructure for turning powerful but flaky AI coders into reliable, long-horizon superh". CTAs: "Install", "Buy lifetime license", "Apply for Founders Forge".

Most likely leak: Copy passes basic thresholds, but adding one proof element (testimonial, case study, or this Proof badge) directly next to the CTA will increase conversion more than any further copy change.
Best next fix: Insert this Proof badge as a trust signal directly next to or inside the primary CTA button — not below the fold, not on a separate page.

How Arena scored this proof — methodology & scope

Method: Live browser agent at desktop 1280x900 + mobile 390x844 via UIAI Engine v23

Coverage: HTML body, public resources, viewport meta, headline, meta description, CTAs, image counts

Dimensions: Clarity 25% / Trust 20% / CTA 25% / Mobile 15% / Friction 15%

Did NOT cover: private dashboards, raw session logs, payment pages, authenticated areas, A/B test, heatmap, real-user behavior

02Powered by UIAI Engine v23

focusa.dev/: 7 checks across 1 viewport — no private pages accessed

URL

https://focusa.dev/

Viewport

desktop 1280x900

Check

headline observed: yes

Check

meta description observed: yes (119 chars)

Check

CTA buttons observed: 7

Check

image count: 0

Check

viewport meta: present

Check

public resources observed: 0

Check

possible failed public resources: 0

Limitations

Public-safe desktop observation only; no private checkout steps were completed, no raw session was kept, and mobile/tablet screenshots require a separate pass.

03Powered by Focusa Ontology

Focusa mapped you as: Install (confidence 0%)

Business / person

Focusa is infrastructure for turning powerful but flaky AI coders into reliable, long-horizon superhuman collaborators.

Likely audience

People motivated by the observed CTA path: Install, Buy lifetime license, Apply for Founders Forge

Primary offer

Install

Primary CTA

Install

Confidence

04Powered by Arena + Focusa

Cold visitor gets 5 things in 10s, blocked by 3

First impression: Kill the chat. Keep the mission.

Understood in 10 seconds:

Three proof cards from real agent work.

Choose the right entry point.

One launch system. Multiple connected offerings.

Agent forgot the mission. Focusa restored it.

Today's agents are powerful, but fragile.

Unclear in 10 seconds:

Whether the promise in the meta description is immediately proven on-page.

Which of the 7 CTAs is the primary purchase path versus supporting navigation.

No screenshot is attached to visually verify the first impression.

Likely visitor question: What proof backs this offer, and where should I click first?

05Powered by UIAI Engine v23

UIAI captured 4 public-safe screenshots — each proves a specific claim

Each piece of evidence is public-safe and linked to a specific claim.

01Headline observationpublic-safe
observationH1: "Kill the chat. Keep the mission."
What it proves: First impression is "Kill the chat. Keep the mission.". If this matches what buyers search for, keep it. If it is only a category name (no buyer, no outcome), visitors who need more specific proof will bounce in <10s.
02Call-to-action inventorypublic-safe
observation7 CTA(s): "Install", "Buy lifetime license", "Apply for Founders Forge", "Review this offer"
What it proves: 7 CTAs found: "Install", "Buy lifetime license", "Apply for Founders Forge". Multiple CTAs dilute focus; promote one as the primary, demote the rest.
03Page structurepublic-safe
observation19 H2 sections, 18 total headings, 0 images, has viewport meta.
What it proves: 19 section title(s) found. No images. Viewport meta present (mobile ready). 19 sections is dense; consider grouping or simplifying to prevent scroll fatigue.
04Browser diagnosticspublic-safe
diagnostics0 resources, 0 possible failures at 1280x900.
What it proves: 0 resources loaded, 0 possible failures. No failed resources detected. The page loaded cleanly from a technical perspective. Viewport: 1280x900.

06Powered by Arena + Focusa

Across 5 dimensions, average 90/100

Every score has a written reason and an Arena benchmark (top quartile, median).

Clarity100/100

H1 "Kill the chat. Keep the mission." gets 50/100 for length and specificity. Section structure: 19 H2s = 90/100. The headline and structure together give a cold visitor a clear first impression.

top 5%median 65 · top Q 85+

Trust100/100

Meta description: Focusa is infrastructure for turning powerful but flaky AI c... (119 chars, optimal 150-180). Meta description and title are well-formed for search discovery.

top 10%median 60 · top Q 80+

CTA100/100

7 CTA(s) found. Verbs: "Install", "Buy lifetime license", "Apply for Founders Forge", "Review this offer". CTAs are clear and use purchase-ready verbs.

bottom 15%median 55 · top Q 80+

Responsive readiness70/100

Viewport meta: present. The page declares viewport meta, so mobile layouts should render correctly on most devices. Verified from desktop pass only; actual mobile/tablet render requires a device-specific check.

top 25%median 55 · top Q 80+

Friction80/100

7 CTA(s) on the page. Multiple CTAs with equal visual weight split the cold visitors attention. Consider promoting one as primary and demoting the rest. Failed resources: 0 may add technical friction (slow loads, broken images).

below medianmedian 55 · top Q 80+

07Powered by UIAI Engine + Focusa

3 buyer-impact findings · 0 ranked High severity

3Headline is specific but could name the buyerLOW

CriteriaBest practice standard for this dimension.

ConditionObserved H1: "Kill the chat. Keep the mission." (32 chars)

CauseFirst-scroll headline sets whether the page reads as "for me" or "generic". 100/100 clarity score.

ConsequenceH1 clarity component +14 points. Visitor bounce rate typically drops 12-20% when the H1 names the buyer specifically.

CorrectiveReplace the H1 with a buyer-named value claim under 12 words.

37 competing CTAs on the pageLOW

CriteriaBest practice standard for this dimension.

ConditionCTAs observed: "Install", "Buy lifetime license", "Apply for Founders Forge", "Review this offer" ...

Cause100/100 CTA score. When the primary action is unclear, conversions split between secondary actions or leave.

ConsequenceCTA score +0-0 points. Primary action click-through typically rises 8-18%.

CorrectivePromote one CTA to a full-width button above the fold; demote the rest to inline text links.

3Meta description is present but 119 charsLOW

CriteriaBest practice standard for this dimension.

ConditionMeta description length 119 chars.

CauseMeta description is what search engines and social cards show. 100/100 trust signal score.

ConsequenceTrust score +0 points. Search click-through rate typically rises 5-12%.

CorrectiveWrite a 150-180 char meta description that names the product, the buyer, and one concrete proof signal.

08Powered by UIAI Engine + Focusa

Priority 1: Top fix: Insert this Proof badge as a trust signal directly next to or inside the primary · expected impact Does not chase fake score gains; converts an already-strong page into a more believable buying moment by showing proof next to the CTA.

Top fix: Insert this Proof badge as a trust signal directly next to or inside the primary

Insert this Proof badge as a trust signal directly next to or inside the primary CTA button — not below the fold, not on a separate page. (Proof #3947.)

Implementation hint: Add a "Proof before purchase" block beside the primary CTA: quote the observed H1, show the 100/100 CTA score, link this Proof, and add one buyer-specific trust line. Estimated 20-40 minutes.

Expected impact: Does not chase fake score gains; converts an already-strong page into a more believable buying moment by showing proof next to the CTA.

09Powered by UIAI Engine + Focusa

Drop-in Suggested rewrite — paste into your CMS, no agency needed

A drop-in replacement you can paste into your CMS today — no rewrites, no agency needed.

Suggested rewrite (other)

BEFORE

Kill the chat. Keep the mission.

AFTER

Your buyer needs public proof that kill the chat. keep the mission. actually works for real buyers. Show it before they scroll.

10Powered by UIAI Engine + Focusa

Three sequenced moves, ordered 24h / 48h / 72h — ship top fix first

1. Place a compact proof card near the primary CTA with score, observed H1, CTA count, and link to this public Proof. Ship within 24 hours.

· effort: medium

Impact: high

2. Name the highest-value buyer segment and one concrete outcome so the strong headline becomes more specific. Ship within 48 hours.

· effort: medium

Impact: high

3. Keep all CTAs, but visually mark the primary purchase path and make secondary actions quieter. Ship within 72 hours.

· effort: medium

Impact: high

11Powered by Focusa Predictions

Arena logged 6 predictions (avg confidence 82%) — will be checked on next proof

If you apply the headline fix, the clarity score should rise 8-15 points.90%

First-scroll clarity is dominated by the headline; observed H1 carries the claim.
→ Rewrite the H1 to a buyer-named value claim.

Adding a 150-180 char meta description should raise the trust score 10-20 points.90%

Meta description is a primary trust signal observed low or missing.
→ Write a meta description that names the product and the buyer.

Promoting one CTA above the fold should raise the CTA score 15-25 points.90%

7 CTAs observed. Conversion lifts when one is visually primary.
→ Pick one CTA, make it full-width above the fold.

Reducing visible competing actions to one primary should reduce measured friction.85%

Multiple CTAs on one page split attention.
→ Demote secondary actions to inline text links.

Proofs with one named number + one named element get shared more often than generic audits.65%

Specific, sourced claims travel further than round numbers.
→ Copy the share card verbatim into a post.

A receipt that names a specific element and a specific fix is more likely to make the buyer confident than one that does not.70%

Specificity is the buyer-confidence multiplier in this product category.
→ Apply the top fix and re-run a receipt to compare.

12Powered by Focusa Ontology

Focusa mapped 5 objects and 2 relations — all evidence-linked

A typed object/relation map of your public presence — every claim linked to evidence. This is what makes Arena's proof deeper than a normal audit.

Objects identified (5)

Objectactive

Page:https://focusa.dev/

Objectactive

Title:645b76a90ff75443fb4d5a73cf4c847a

Objectactive

CTA:7

Objectactive

ProofSignal:Redacted

Objectactive

Meta:7468a0b77621a8900ecb94ebd91dd6a5

Relations detected (2)

relates to →
relates to →

Conflicting actions (3)

Buy receiptJoin ForgeBook setup call

✓ Verified claims (3)

Page loaded successfully at https://focusa.dev/
Headline element present: yes
7 CTA button(s) detected on the page.

? Uncertain claims (1)

Meta description was observed but may be reworded by the platform.

13Powered by Focusa Metacognition

Focusa captured 1 reusable lesson — future proofs will be sharper

A reusable lesson extracted from this proof — captured so future proofs get sharper, not just longer.

Retrieved lesson

A receipt feels weighty when each section names a specific element observed on the page, not a generic copy template.

Applied here

This Proof observed 19 H2 sections, 7 CTAs, and a 32-char H1. Each section is grounded in those observations.

New lesson captured

The buyer values the top fix most when it names the exact element (not "improve copy" but "rewrite the H1 to name the buyer").

Next receipt guidance

Capture the actual screenshot of the buyer page in evidence[0] and reference it inline in the verdict and findings.

16Powered by UIAI Engine + Focusa

Want fresh scores? Get a new Proof.

This Proof is compiled once and locked in time — frozen at 2026-08-04 10:59 CST. Scores, predictions, and lessons do not refresh on later visits. If you want fresh numbers for an updated page or a new campaign, purchase a new Proof.

Site or Profile Proof from $49 · Rival Proof from $149

Get My New Proof

Site Proof #3947

Your page scores 90/100 · risk low · biggest leak: Copy passes basic thresholds, but adding one proof element (testimonial, case study, or...

Your headline "Kill the chat. Keep the mission." communicates a clear value; the lift is in proof near the CTA.

focusa.dev/: 7 checks across 1 viewport — no private pages accessed

Focusa mapped you as: Install (confidence 0%)

Cold visitor gets 5 things in 10s, blocked by 3

UIAI captured 4 public-safe screenshots — each proves a specific claim

Across 5 dimensions, average 90/100

3 buyer-impact findings · 0 ranked High severity

Priority 1: Top fix: Insert this Proof badge as a trust signal directly next to or inside the primary · expected impact Does not chase fake score gains; converts an already-strong page into a more believable buying moment by showing proof next to the CTA.

Top fix: Insert this Proof badge as a trust signal directly next to or inside the primary

Drop-in Suggested rewrite — paste into your CMS, no agency needed

Three sequenced moves, ordered 24h / 48h / 72h — ship top fix first

Arena logged 6 predictions (avg confidence 82%) — will be checked on next proof

Focusa mapped 5 objects and 2 relations — all evidence-linked

Objects identified (5)

Relations detected (2)

Conflicting actions (3)

✓ Verified claims (3)

? Uncertain claims (1)

Focusa captured 1 reusable lesson — future proofs will be sharper

Post this proof to LinkedIn, X, Reddit — one click, no copy/paste

Pre-filled LinkedIn post — click Copy if you prefer manual posting

Want fresh scores? Get a new Proof.