Additional PDFTEQ

[kureshi449621-ui]

Additional Requirements / Suggestions for BeyondSEO

Please improve BeyondSEO as a reusable, production-grade SEO engineering skill for ChatGPT Work.

1. Full Website Audit

• Crawl the complete website, not only the homepage.
• Discover PHP/HTML pages, dynamic routes, sitemap URLs, canonical URLs, robots.txt, images, PDFs, and important assets.
• Detect orphan pages, broken internal links, redirect chains, duplicate URLs, missing pages, and indexing issues.

2. Technical SEO

• Check title tags, meta descriptions, H1-H6 hierarchy, canonical tags, robots directives, XML sitemap, robots.txt, hreflang, Open Graph, Twitter Cards, structured data, pagination, redirects, HTTP status codes, and URL structure.
• Detect conflicting or duplicate SEO signals.
• Validate sitemap URLs against actual HTTP responses.

3. Modern Core Web Vitals

• Use current Core Web Vitals, especially LCP, INP, and CLS.
• Do not rely on obsolete FID terminology.
• Identify render-blocking resources, excessive JavaScript, unused CSS, image loading problems, font issues, TTFB, caching, and third-party scripts.
• Provide practical code-level fixes where possible.

4. Structured Data

• Automatically detect and validate Organization, WebSite, BreadcrumbList, Product, Article, FAQPage, LocalBusiness, WebPage and other relevant Schema.org types.
• Detect conflicting, incomplete, duplicated, or invalid JSON-LD.
• Recommend the correct schema based on actual page content.

5. E-E-A-T

• Audit author information, company identity, contact information, credentials, policies, references, trust signals, and first-hand experience signals.
• Identify missing trust elements without inventing credentials or claims.

6. AI Search / AEO / GEO

• Optimize pages for AI search engines and answer engines.
• Identify question-based search opportunities.
• Improve concise answer blocks, definitions, FAQs, entity clarity, citations/references, and semantic coverage.
• Clearly distinguish factual content from claims.

7. Keyword & Search Intent Analysis

• Analyze primary, secondary, semantic, long-tail, and topical keywords.
• Classify search intent.
• Detect keyword cannibalization.
• Identify content gaps and internal-link opportunities.
• Avoid keyword stuffing.

8. Internal Linking

• Build an internal-link graph.
• Identify orphan pages and weakly linked pages.
• Recommend contextual anchor text.
• Detect excessive or repetitive anchor text.
• Suggest links based on topical relevance rather than arbitrary keyword matching.

9. Image SEO

• Detect missing or poor ALT text.
• Check image dimensions, aspect ratios, modern formats such as WebP/AVIF, lazy loading, fetch priority, responsive images, and image compression.
• Detect layout-shifting images caused by missing width/height attributes.

10. International SEO

• Correctly audit multilingual websites.
• Validate hreflang implementation, language-region codes, canonical relationships, and alternate URLs.
• Do not remove language support merely because a language is associated with a particular country or market.

11. E-commerce / Product SEO

• Audit product titles, descriptions, specifications, images, Product schema, availability, pricing where applicable, SKU/GTIN fields, breadcrumbs, and product-category relationships.
• Detect thin product pages and duplicate product descriptions.

12. Content Quality

• Detect thin, duplicate, outdated, generic, AI-like, or low-value content.
• Suggest improvements while preserving the website's actual facts and business positioning.
• Never invent statistics, certifications, customers, locations, experience, or claims.

13. Crawler Reliability

• Provide separate readiness checks for:
  • HTTP/network access
  • DNS
  • robots.txt
  • sitemap
  • browser/Playwright
  • JavaScript rendering
  • Python runtime
  • filesystem permissions
• Do not report the crawler as "ready" if only a format/configuration check passed.

14. Doctor Command
    The doctor command should clearly report:

• PASS
• WARN
• FAIL
• BLOCKED

For every failure, show:

• exact command
• exact error
• likely cause
• required fix
• whether the failure blocks crawling

15. Safe Code Modification
    When code modification is supported:

• Create a backup before modifying files.
• Make targeted changes only.
• Preserve existing design and functionality.
• Never modify unrelated files.
• Run syntax/lint checks after modifications.
• Run a second SEO crawl after modifications.
• Show a before/after summary.

16. PHP Websites
    Support PHP websites properly:

• Inspect PHP templates/controllers/views.
• Detect dynamic routes.
• Understand database-driven sitemap/country/product pages where possible.
• Never assume a PHP site is WordPress.
• Run php -l against modified PHP files.

17. Hostinger / Shared Hosting
    Include practical checks for:

• LiteSpeed Cache
• .htaccess
• PHP versions
• compression
• browser caching
• redirects
• HTTPS
• canonical host
• trailing slash consistency
• sitemap/robots accessibility.

18. Reporting
    Generate:

• Executive Summary
• Critical Issues
• Technical SEO
• On-Page SEO
• Content
• Schema
• Performance
• Internal Links
• Crawlability/Indexability
• E-E-A-T
• AEO/GEO
• Prioritized Fix Plan
• Verification Results

Every issue should include:

• URL/file
• severity
• evidence
• explanation
• recommended fix
• verification method

19. Backlink & Off-Page SEO
    Use the included backlink catalog to identify relevant, legitimate opportunities.

• Categorize opportunities by type and relevance.
• Avoid spam, PBNs, automated link farms, paid-link schemes, and unsafe domains.
• Do not recommend mass automated submissions.
• Track prospect, URL, domain, relevance, contact method, status, and target anchor/context.

20. Reusable Skill Design
    The complete BeyondSEO skill should remain reusable across different websites and projects.
    It should not hard-code assumptions about:

• WordPress
• one programming language
• one hosting provider
• one country
• one business type.

21. No False Success
    Most importantly:
    Never report "installed", "ready", "healthy", "fixed", or "100% complete" unless the corresponding operation has actually been executed and verified.

A documentation check, syntax check, configuration check, or format check alone must never be treated as proof of successful crawler installation or browser readiness.

22. Evidence-Based Results
    For every important finding, provide evidence from:

• crawler output
• HTTP response
• source code
• browser rendering
• sitemap
• robots.txt
• structured-data validation
  or another clearly identified source.

If something could not be tested because of network, permission, runtime, browser, or environment limitations, explicitly mark it as BLOCKED rather than guessing.

Please incorporate these requirements into the reusable BeyondSEO Work skill, its playbooks, crawler doctor checks, reporting workflow, and verification process where technically appropriate.
Technical Team: PDFTEQ

[beyondtahir]

Thank you to the PDFTEQ team for putting this together. We reviewed your suggestions and used them to strengthen BeyondSEO 2.9.0.
Here’s what we’ve added:

• Flexible website audits: check the homepage, a standard sample of 15 priority pages, specific URLs or a larger selection. Saved audits can continue with another batch.
• Clear coverage: reports distinguish inspected, partially captured, blocked and remaining pages. An unsuccessful capture is not treated as evidence of missing content.
• Deeper technical checks: sitemap URLs checked against observed responses, language-page relationships, duplicate descriptions, duplicate or conflicting JSON-LD, and image attributes.
• More useful diagnostics: separate checks for Python, filesystem access and browser JavaScript, with optional live checks for DNS, website access, robots rules and sitemaps.
• Safer changes: PHP syntax checks before staging and applying supported PHP edits, alongside clearer guidance for backups, fresh crawls, layout verification and rollback.
  Your point about “no false success” was particularly important. Installing the skill, having a working browser, reaching a website and verifying a fix are separate outcomes—and should be reported separately.
  For validation, 249 automated tests passed locally, with one platform-specific test skipped. A separate local trial checked page selection, applying a change, desktop/mobile layout preservation and rollback.
  This release doesn’t implement every item in the proposal. Full Core Web Vitals measurement and comprehensive Schema.org validation remain beyond these new checks. Real PHP lint execution and fresh imports into every assistant also remain untested for this release.
  You can read the update here:
  BeyondSEO 2.9.0 — Know your website better. Fix it with confidence.
  Thanks again for helping us make BeyondSEO more useful and accountable.