Ranked by what each change is worth on the composite — the gap to the next rung of the ladder times the dimension’s weight. Computed from the scores, not written by the model; the model’s sentence under each is the action.
Use a concrete product noun in the headline, e.g., "Isolated sandbox environments for running AI agents".
Use a direct, unambiguous CTA aligned with the product: e.g., "Start Free Sandbox Trial".
All three together: 79 → 86. Every dimension one rung up: 86 (grade A). Ceiling is 95.
Built only from facts already on the page — not the company’s words. Shown so the gap between what was written and what could have been written is concrete rather than a number.
Five alternative headlines in five angles, a subhead, a call-to-action line, and a line-by-line edit for each of the five dimensions — written only from facts already on agent37.com, gated like a rating (anything with a struck word is dropped, not sold). Delivered on a private page in about a minute.
Five dimensions on one ladder — 5 · 20 · 35 · 50 · 65 · 80 · 95 — weighted clarity 30%, specificity 20%, jargon 20%, proof 15%, call to action 15%. For each: the rung the page matched, the words that drove it, and what would reach the next rung. Full rubric →
Decent clarity but the headline uses an abstract term rather than a concrete product noun.
Host any agent harness | Agent 37
Next rungUse a concrete product noun in the headline, e.g., "Isolated sandbox environments for running AI agents".
The page includes a strong numeric outcome (percent cost reduction).
at up to 95 % lower cost than most sandbox providers.
The text uses few abstract terms in the main claim, and relies on concrete nouns like 'sandbox' and 'agents'.
Scored from the page as a whole; no single line was cited.
Multiple explicit numbers and named customers anchor the claims.
Powering agents for over 100 companies
Multiple CTAs exist (Start Free, Book a call) that loosely connect to the claim.
Start Free Talk to the founder
Next rungUse a direct, unambiguous CTA aligned with the product: e.g., "Start Free Sandbox Trial".
Host any agent harness | Agent 37Shows the main headline, which is vague and not a clear product noun.
at up to 95 % lower cost than most sandbox providers.Numerical claim tied to a concrete outcome (cost).
Host any agent harness | Agent 37 Skip to content Backed by Y Combinator Host any agent harness. Hermes OpenClaw Claude Code Codex OpenCode Grok Pi Custom harness Give every customer an agent in its own isolated sandbox , at up to 95 % lower cost than most sandbox providers. Start Free Talk to the founder $5 free credit Paste this into Claude Code, Cursor, or any coding agent. Read the docs first: https://www.agent37.com/docs/llms-full.txt Build me a real estate assistant using Hermes. Read the docs first: https://www.agent37.com/docs/llms-full.txt Build me a social media agent using OpenClaw. Read the docs first: https://www.agent37.com/docs/llms-full.txt Build me a code-review agent using the Claude Agent SDK. Read the docs first: https://www.agent37.com/docs/llms-full.txt Build me a coding assistant using OpenCode with DeepSeek Flash as the default model. Read the docs first: https://www.agent37.com/docs/llms-full.txt Build me a real estate assistant using Hermes. Powering agents for over 100 companies ApolloClaw Boon Vision Perflo LeniHealth ETEC+ Lionox Future Bridge Solutions Gedi Foundation Filament Contractor Scale Sareli Systems Pricing 95% lower cost than most sandbox providers Find a cheaper sandbox? We’ll refund the difference. 2 vCPU 4 GB 4 vCPU 8 GB 8 vCPU 16 GB 4 GB persistent disk Hours used 7 % · 57 hrs 1 hour Always running Monthly estimate $0.37 Compute $0.34 Disk $0.03 Monthly usage comparison Sources vs Agent37 Agent37 $0.37 1× Hetzner CX23 · EU $6.49 17.
Strongest: Specificity 95. Weakest: Clarity 50. The 2 passes agreed within 1 points. The band 74–84 is where a re-score of this unchanged page would most likely land: on 42 dry re-scores (12 Oct 2026) the median move was 6 points and one in four moved more than 10. A grade only changes when a new score clears the boundary by 5.
No word in the headline itself was judged empty. That is rare: 13 of 134 companies in the index manage it.
Static SVG. No JavaScript, no tracking.
Scored from the live homepage
on 2026-10-13 by gpt-5-nano,
2 independent passes, median taken.
Page snapshot 843be44f583a.
Rubric 7329c240 (current; what it says).
No person reviewed or edited this score. If the rubric and the score disagree, the rubric wins
and the page is re-scored — not hand-corrected.
Re-scored when the homepage text changes.