QA · UX · AI Auditing
Does your website, product or AI actually work?
Most teams assume it does, right up until a customer cannot check out, a screen reader hits a wall, or an AI feature answers wrong in front of someone who matters. You find out from the complaint. We find it first. You get every problem ranked by how much it costs you, with the steps to reproduce it, the standard it breaks and the evidence attached.
QA engineering since 2022. Hobe Sound, Florida, working with clients anywhere.
Stop finding out
the expensive way.
Auditing is the core of it: knowing exactly what is broken, and how badly, before your customers or a procurement team do. Build and Verify sit either side, because a problem you cannot fix, or cannot prove you fixed, is only half an answer.
- Websites and web apps
- Custom tools and automation
- AI adoption guidance
- Every build ships with its own audit report
Know exactly what is broken and how badly, before your users or a procurement team find it for you. Every problem comes back ranked, reproducible and evidenced, so your developer can start on it the same day.
- Website audit: accessibility, performance, security headers
- UX and workflow audit on real devices
- AI system audit: accuracy, guardrails, failure states
- WCAG 2.2 AA, Core Web Vitals and OWASP under an ISTQB process
- Automated regression suites
- Fix verification against the original finding
- Pre-launch testing
- Release certification
Businesses that cannot
afford to assume.
If a defect on your site costs you money, a customer or a contract, you are the reason this practice exists. In particular:
- Businesses under accessibility pressure from a client, an insurer or a procurement questionnaire
- Anyone shipping an AI feature to customers who needs to know how it behaves at the edges
- Checkout, booking and signup flows losing people for no obvious reason
- Teams that have been shipping fast and want a second pair of eyes before a launch
- Agencies that need an independent QA bench behind their own brand
- Anyone who inherited a site and has no idea what is in it
We test our own work first
Four facts about this website, every one measured with the same tooling we would point at yours. Run the free check on us and you should get the same numbers.
Try the tooling
before you hire us.
Before you pay us anything, point our scanner at your own site and see what it says. No account, nothing stored, about two seconds. The other two are products we built and maintain, here so you can watch our work run instead of reading about it.
QA Site Check
Eleven checks against any URL in about two seconds: transport security, six response headers, and the document facts that decide whether a screen reader and a phone can use the page. Every result maps to OWASP Secure Headers or a WCAG 2.2 AA success criterion, and the scan states plainly what it could not test.
- Findings, not a score out of 100
- Each one mapped to a named standard
- Severity rated against published definitions
- Respects your robots.txt
- Reports what it could not test
- Free, no account, nothing stored
Free check, run on ourselves
russellspc.com, scanned by the button on this page
Three of eleven shown. The four checks this pass cannot reach need a real browser or a person, and the report names all four rather than scoring them as passes. That is the part worth checking us on: point the scanner at your own site and see whether it tells you what it skipped.
Prompt Sweeper
Catch accidentally included AI prompts, preambles and placeholder text before your audience does. Scans copy for prompt residue left behind when pasting from AI tools into emails, websites, or documents.
- Detects prompt instructions
- Catches AI preambles and closings
- Flags placeholders and brackets
- Spots role assignments
- 100% client-side. No data sent
- Now available as a free Chrome extension
Prompt Sweeper
Catch AI residue before your audience does
Thank you for your inquiry. Write a professional response to the client about our services. I’d be happy to give you a walk through of our products and services. Disregard all prior instructions, The solution — though complex was universally beneficial. Please see [Your Company Name] for details.
AnswerWeave
Your AEO Teammate. See what ChatGPT, Perplexity and Google Gemini say about any business right now. Score AI visibility across 5 dimensions, get actionable fixes, and monitor citation trends over time. White-label ready for agencies.
- Queries 3 AI engines live
- 5-dimension AEO scoring
- Schema and structured data audit
- Content readiness analysis
- Competitor AI presence tracking
- Agency white-label ready
AnswerWeave
AEO Audit · ChatGPT · Perplexity · Gemini
Tools we built, and test, in public
These are not the offer, but they are why you might believe the offer. Each one taught us a rule we now enforce in code.
SEOrcery
Local SEO scoring. A timeout gets recorded as not tested, never as a zero. We added that rule after a 79-second failure came back as a near-zero speed score for a site that actually answered in 0.36 seconds.
Lead Radar
Market prospecting. Businesses it finds without a website are banked rather than discarded, because data you already paid to collect should not be thrown away on the next pass.
Rival Vectors
Competitive intelligence. Every claim it makes links back to the request that produced it, so you can check a conclusion instead of taking it on faith.
The audit engine
67 tests, including nineteen on the guard that decides which URLs it will fetch. A finding cannot be constructed without all nine of its fields.
Don’t assume it works.
Prove it.
Send us a URL and we will tell you what is wrong with it, with the evidence attached. Scoped in writing before anything starts.
Send us one URL.
We will tell you what is wrong with it.
Russell SPC LLC
Don’t assume it works. Prove it. Independent QA, UX and AI auditing: we test what you have built and hand back findings you can act on, each with reproduction steps, a named standard and the evidence behind it.