- Home
- Skills
- Testing & Debugging
- KeurSmid: 8-Agent Website QA Audit
More screenshots
Works with the AI Agents you already use
KeurSmid: 8-Agent Website QA Audit
Parallel multi-agent site QA for link checking, SEO, visual UX, and i18n audits with automated HTML reporting.
$10
KeurSmid: 8-Agent Website QA Audit
Example session with this skill installed
/keursmid
URL: https://books.toscrape.com/
Pages: /, /catalogue/category/books/travel_2/, /catalogue/category/books/mystery_3/, /catalogue/a-light-in-the-attic_1000/, /catalogue/its-only-the-himalayas_981/, /catalogue/page-2.html
- Read your context and instructions
- Compiled the keursmid
- Generated the document
KeurSmid run complete for books.toscrape.com: 7 of 8 agents ran (Translation Check skipped: single-locale site). Report: tests/keursmid-report-2026-10-13-v2.html (opened in your browser).
167 passed (124 of them link checks) · 17 failed · 24 warnings · 51% pass rate (link checks excluded).
Top actions
- FAIL: Add Open Graph and Twitter card tags (og:title, og:description, og:image, og:url, twitter:card) to every page: all 6 audited pages share as a bare link with no image (Agent 7)
- FAIL: Give product detail pages an Add to basket button: /catalogue/a-light-in-the-attic_1000/ has zero CTAs and is a dead end (Agent 3 confirms no buy button on the decision page) (Agent 8)
- FAIL: Add a self-referencing canonical link to every page: missing on all 6 (Agent 6)
- FAIL: Fill the empty meta description on the home, Travel, Mystery and catalogue page-2 listings (Agent 6)
- WARN: Load jQuery 1.9.1 over HTTPS: the http://ajax.googleapis.com script is blocked as mixed content on every page, so any jQuery behaviour is dead (Agent 1)
Per-agent results
| Agent | Pass | Warn | Fail | Notes |
|---|---|---|---|---|
| 1 Pages | 0 | 6 | 0 | All 6 pages return 200; every page warns on the blocked mixed-content jQuery script |
| 2 Translation Check | Skipped: single-locale site | |||
| 3 Visual | 17 findings (4 high) from 12 desktop and mobile screenshots | |||
| 4 Competitors | 2 demo bookstores benchmarked, 7 gaps | |||
| 5 Links | 124 | 0 | 0 | Internal, external and anchor links all resolve |
| 6 SEO | 38 | 18 | 10 | Canonical missing on all pages; empty meta descriptions on 4 |
| 7 Social | 0 | 0 | 6 | No Open Graph or Twitter tags on any page |
| 8 Conversion | 5 | 0 | 1 | Product page has no purchase CTA |
The HTML report shows each finding with its page, evidence and screenshot; the four images in this listing are from this run's report.
keursmid-report-2026-10-13-v2.html
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Ship web projects with confidence by catching broken links, missing SEO and social tags, and UX dead ends before your users do. KeurSmid automates the tedious pre-launch QA process by dispatching up to eight specialized agents in parallel to audit your site (agents that do not apply, such as translations on a single-language site, are skipped and reported). It handles everything from technical link validation to competitor benchmarking, consolidating all findings into a single, actionable HTML report.
What it does
- Multi-agent QA dispatches parallel agents to check pages, translations, visual UX, and SEO.
- Automated link checking verifies internal and external URLs including anchor links for up to 500 targets.
- Cross-locale validation detects i18n routing issues and ensures H1 tags match the expected language.
- Visual UX review captures desktop and mobile screenshots and reviews them for layout, readability and UX problems (a critique, not a pixel-diff against a baseline).
- Competitor benchmarking discovers market rivals and compares design, trust signals, and CTAs.
- Social preview audit validates Open Graph and Twitter cards to ensure brand consistency on social feeds.
How it works
- Initialize by running
/keursmidto trigger project auto-detection for Next.js, Nuxt, or Astro. - Pre-flight checks confirm the site is reachable and verify Playwright dependencies are installed.
- Dispatch launches specialized agents to collect structured JSON data from your local or staging URL.
- Compile runs the bundled report builder: one self-contained HTML report with every finding, the top actions, and a pass rate that excludes link checks so they cannot inflate it.
Frameworks & tools
This skill utilizes Node.js and Playwright for browser automation. It features native auto-detection for Next.js, Nuxt, and Astro, but supports any web framework via manual URL configuration.
Why this beats prompting it yourself
Writing reliable browser automation scripts for eight different domains is time-consuming and prone to flake. This skill provides a tested orchestration layer that handles parallel execution, dependency management, and structured data synthesis that a single prompt cannot replicate.
Use cases
- Pre-launch audits to catch 404s and missing meta tags before a production deployment.
- Periodic health checks after merging major dependency updates or UI refactors.
- Competitor analysis to quickly benchmark your site's conversion flow against market rivals.
- i18n verification ensuring localized routes display the correct content and headers.
Known limitations
Does not provide deep performance benchmarking like Lighthouse or full WCAG accessibility audits. Link checking is capped at 500 links. Manual confirmation is required for competitor lists.
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scanned
Verified clean 7 days ago
- Passed all security checks, Safe to install
Requires