Sydney Business Web / Engineering research
Research & Publications
Open-access engineering reports and software records by Keith Rowley, Sydney Business Web. Explore documented methods, production investigations and technical releases across business identity, AI visibility, crawler retrieval and website infrastructure. Every record below has a direct DOI link—no Zenodo search is required.
Engineering reports & case studies
Latest first · 2026
Cloudflare HTML Caching for WordPress: Fault Isolation, Safe Edge-Cache Boundaries and a Device-Specific Regression
A production WordPress caching study documenting regional HTML delivery latency, narrowly scoped edge-cache rules, measured TTFB and LCP improvements, and a subsequent device-specific cache regression.
Cloudflare Markdown for Agents: JSON-LD Fidelity — A Production Investigation of HTML Conversion, Content Negotiation and Structured-Data Integrity
A production HTML-to-Markdown investigation reporting six matched JSON-LD structures and independently reproducing a multi-object JSON-fence parsing issue from preserved Markdown evidence.
Version-Dependent Agentic Browsing Audits: A Controlled Lighthouse 13.5–13.6 Investigation of ARD Discovery
A controlled comparison of preserved Lighthouse 13.5.0 and 13.6.0 JSON audits showing ARD applicability changed from Not Applicable (3/3) to Pass (4/4), with unchanged category score.
DOI-Backed Research Evidence Mapping: A First-Party Publication Index and Entity-Graph Implementation
A documented snapshot of the initial 15 DOI-linked records, verifying first-party publication-list consistency and canonical author/business identity references without claiming external AI uptake.
Automated Catalogue Requests and PHP-FPM Saturation: A 17,000-Product WooCommerce Case Study
A production VPS investigation documenting confirmed PHP-FPM saturation, suspicious automated catalogue requests, targeted ModSecurity controls and bounded follow-up verification.
AI Baby Gorilla: From Single-Page Identity Diagnostics to Public AI-Agent Execution
Engineering validation of a single-page JSON-LD checker, public API discovery, access controls and operator-tested agent execution.
AI Identity Diagnostic: A Verified Person Identity Consolidation Case Study
A before-and-after case study of canonical Person identity repair: four Person clusters reduced to two across 274 analysed pages.
Measuring Verified AI Crawler Retrieval: A Server-Side Observability Methodology
A server-side measurement approach for qualifying machine requests, separating verified retrieval evidence from search ranking and citation claims.
Schema Gorilla: Whole-Site Entity Graph Analysis and Verified Identity Remediation
Whole-site structured-data graph analysis, defect correction and fresh-run verification against an implemented ruleset.
Agentic Browsing Readiness: Diagnosing and Remediating Machine-Readable Website Defects
Investigation of structural website defects that can affect automated browser interaction despite a visually functioning page.
HTTP 429 Rate Limiting of GPTBot-Labelled Requests
Evidence-led investigation of crawler-labelled HTTP 429 responses across Cloudflare, Apache and Imunify360.
HTTP 418: An Origin-Layer ModSecurity Block Affecting AI Crawler Retrieval
A controlled diagnosis of an HTTP 418 access failure involving ClaudeBot-labelled requests and an origin-layer security rule.
Backing Up Cloudflare Workers, D1 Databases and Managed VPS Infrastructure
A technical field report on preserving Worker sources, exporting D1 data and maintaining recoverable hosting infrastructure.
The AI Credibility Footprint: An Engineering Framework for Business Evidence and AI Visibility
A framework for assessing the consistency, provenance and external corroboration of machine-readable business evidence.
The Intelligent Entity Skeleton: An Engineering Framework for AI Visibility and Business Identity
A technical framework for connecting first-party business identity, structured entities and supporting evidence across a website.
Software & technical releases
Four original SBW software records
Schema Gorilla
Whole-site structured-data entity graph analysis, relationship discovery and identity fragmentation diagnostics.
AI Observatory
Server-side monitoring and reporting of qualifying AI/search crawler HTTP retrieval activity.
AI Identity Diagnostic
Interpretation of whole-site entity graph findings into bounded identity-assessment states and remediation guidance.
AI Baby Gorilla
Single-page JSON-LD entity-connectivity checker with an externally callable diagnostic interface.