Brutal Audit
LLM Gateway
Powered by LLM Gateway

A brutally honest auditof your site, in 60 seconds.

Pick an angle. SEO, AI search, copy, psychology. Get a real score, real findings, and ready-to-paste prompts to fix it. No fluff.

1
Connect
2
Target
3
Angle
4
Verdict
1

Connect your LLM Gateway

Paste an API key. We'll use it to run the audit on your model of choice.

Don't have one? Sign up free at llmgateway.io.

2

What are we auditing?

Drop any public URL. We fetch it server-side and analyze the static HTML.

3

Pick your audit angle

One skill, one focused audit. You can re-run with a different angle after.

8 angles

Pick your audit angle

One angle per run. Each one uses a different framework with its own scoring criteria. Re-run on the same URL with a different angle to triangulate.

SEO Audit

#

Why isn't Google ranking you?

Technical + on-page audit. Crawlability, indexation, Core Web Vitals, title/meta/H1, internal linking, canonicals.

AI Search (AEO/GEO)

#

Are you visible to ChatGPT & AI Overviews?

Audit for visibility in ChatGPT, Perplexity, AI Overviews. Chunk-friendly content, citations, E-E-A-T, llms.txt.

Copywriting

#

Does your copy actually sell?

Hero, value prop, CTAs, social proof, objection handling. Clarity > cleverness.

Copy Editing

#

Where's your copy leaking trust?

Tighten, sharpen, fix tone drift. Cut filler. Spot AI-tells (em dashes, 'delve', 'in today's fast-paced world').

Marketing Ideas

#

Untapped angles for growth.

Generate distribution and growth ideas based on positioning, audience, and signal in the page.

Marketing Psychology

#

Which mental models is the page missing?

Audit through cognitive bias + behavioral science. Social proof, loss aversion, anchoring, scarcity, IKEA effect.

Programmatic SEO

#

Can you scale to thousands of pages?

Spot template-able page patterns, head-vs-tail keywords, datasets you could turn into ranking pages.

Schema Markup

#

What rich results are you missing?

JSON-LD audit. Organization, WebSite, Article, Product, FAQ, BreadcrumbList — what's there, what's missing.

60 seconds, end to end

How it works

  1. Step 01

    Connect your key

    Paste your LLM Gateway API key. We ping the gateway with a real auth check to confirm it works.

  2. Step 02

    Drop a URL

    Any public page. We fetch it server-side and pull title, meta, headings, links, images, and JSON-LD.

  3. Step 03

    Pick an angle

    Choose from the 8 audit angles. Each one applies a different framework with its own scoring criteria.

  4. Step 04

    Get the verdict

    Score, grade, findings with evidence, and copy-paste prompts you drop into Claude or ChatGPT to fix each issue.

Output

What a report looks like

Every audit returns a numeric score, a letter grade, a verdict headline, an executive summary, a list of strengths, an evidence-grounded list of findings, and a stack of recommended actions. Each action ships with a ready-to-paste prompt — the actual current copy from your page is baked in, so you can drop it into Claude or Cursor without filling anything in.

  • 0–100 score with letter grade (A through F)
  • Findings tagged with severity (high / medium / low) plus the exact evidence from your page
  • Self-contained prompts — copy one or copy them all
  • A one-line share tagline you can drop on X
Brutal Audit social preview showing a brutally honest, AI-powered site audit produced in 60 seconds across SEO, AI search, copywriting, and psychology angles
Bring your own key

Why LLM Gateway

Your spend, your model

You pick the model. GPT-4o-mini for cheap and fast, Claude Sonnet for sharper output, free models for zero spend. We don't markup or middleman.

300+ models, one key

LLM Gateway gives you a single API key for every major provider. Swap models without touching code by changing the LLMGATEWAY_AUDIT_MODEL env var.

Pennies per audit

A full audit with gpt-4o-mini costs under 2 cents. New users get free credits to start, so the first audit is on the house.

Get a free LLM Gateway keyBrowse supported models