Overall AI readiness score
A combined result across all audit signals.
Excellent
AI readiness audit: 8/3/2026Public technical data
Overall AI readiness assessment
How ready jprokay.com is for AI systems
Julian Prokay's Blog (jprokay.com) received an AI-readiness score of 80/100 in an automated technical audit. llms.txt was accessible, llms-full.txt was not found, and ai.txt was not found. The robots.txt analysis found 0 explicitly allowed and 0 blocked AI crawlers, with 1 declared sitemap. Homepage markup completeness was 60%; no Schema.org types were detected and 6 OpenGraph tags were detected. Results reflect the public site response observed on 2026-08-03T00:06:03.684Z.
Audit context
llms.txt is accessible and contains 1,463 tokens. No accessible llms-full.txt was detected, so deeper context still has to be assembled from regular pages. No separate ai.txt policy was detected; it is optional, but can clarify training, retrieval, and attribution preferences.
robots.txt is available. 11 of 11 tracked AI bots are not blocked. Declared sitemaps: 1. No Schema.org type was detected on the homepage. 6 OpenGraph tags were found and markup completeness is 60%, leaving more entity interpretation to crawlers.
The mobile Lighthouse profile adds Performance 92/100, Accessibility 100/100, Best Practices 92/100, SEO 100/100, and experimental Agentic Browsing 100/100. These signals have a limited weight: they complement rather than replace llms.txt, robots.txt, and structured-data checks.
Confirmed strengths
- Accessible llms.txt
- 1 sitemap declaration
- Strong Google agentic signals
Priority improvements
- 1Add llms-full.txt for richer first-party context.
- 2Add JSON-LD for the organisation, website, and core entities.
- 3Complete OpenGraph and canonical homepage metadata.
llms.txt tokens1,463
llms-full.txt tokens—
ai.txt
sitemap.xml
Google Lighthouse technical profile
A mobile Lighthouse measurement. Google’s experimental Agentic Browsing category is explained separately and does not replace the broader llmsmap AI-readiness score.
92
Performance
100
Accessibility
92
Best Practices
100
Technical SEO
100
Agentic Browsing
What these results mean
The mobile page renders efficiently; its largest visible content block appeared in 2.7 s.
Accessibility scored 100/100, Best Practices 92/100, and technical SEO 100/100. The experimental Agentic Browsing category scored 100/100. It measures signals Google currently tests for software agents and is shown separately from the llmsmap AI-readiness score.
1
Free the main thread and accelerate the first view
Split long JavaScript tasks, defer non-critical scripts and styles, and shorten blocking request chains. This helps the primary content appear sooner and makes controls usable earlier.
FCP2.7 s
First content
LCP2.7 s
Main content
CLS0
Layout stability
TBT0 ms
Blocking time
SI2.7 s
Visual speed
Metric glossary
- FCP · First content
- When the first text or image appeared on screen.
- LCP · Main content
- When the largest visible element in the first viewport rendered.
- CLS · Layout stability
- How much content shifted unexpectedly while loading; lower is better.
- TBT · Blocking time
- How long the browser main thread could not respond quickly to input.
- SI · Visual speed
- How quickly the visible viewport filled with content.
8/3/2026Lighthouse 13.4.1Mobile profile
AI readiness checks
Machine-readable files, crawler policy, discovery, and homepage markup.
llms-full.txt
Full version was not found
ai.txt
ai.txt file was not found
Sitemap in robots.txt1
1 sitemap found
Schema.org (JSON-LD)
Schema.org markup was not found on the homepage
OpenGraph60%
6 OG tags found
AI bot access
Based on robots.txt analysis
GPTBotNot mentioned
OAI-SearchBotNot mentioned
ChatGPT-UserNot mentioned
Google-ExtendedNot mentioned
ClaudeBotNot mentioned
Claude-SearchBotNot mentioned
Claude-UserNot mentioned
BytespiderNot mentioned
CCBotNot mentioned
PerplexityBotNot mentioned
Perplexity-UserNot mentioned
Sitemaps
Declared discovery routes for crawlers and agents.
OpenGraph tags
Social preview metadata found on the homepage.
og:titleJulian Prokay — Principal Engineerog:descriptionPrincipal engineer with 15 years of experience architecting systems, building teams, and deploying infrastructure.og:typewebsiteog:urlhttps://jprokay.comog:site_namejprokay# Julian Prokay > Principal engineer with 15 years of experience architecting systems, building teams, and deploying infrastructure. Writing about AI-assisted development, software engineering, and the systems around both. I'm Julian Prokay. I currently work on agentic insurance at Counterpart and write about the practical side of building software with agents: where they help, where they make a mess, and how the engineer's job changes when writing the code is no longer the bottleneck. ## Consulting I help startups and scaleups with engineering hiring and practical AI-assisted development using tools such as Pi and Claude Code. That can mean designing an interview loop, reviewing an agent workflow, training a team, or putting guardrails around agentic development. I am not selling a generic "AI transformation." Contact: [LinkedIn](https://www.linkedin.com/in/julian-prokay/) Structured details: [profile.json](https://jprokay.com/api/profile.json) ## Top Posts - ["Claude Code Alternatives: Uncoupling from Anthropic with Pi Agent"](https://jprokay.com/post/020-uncoupling-from-anthropic): Strategies for model independence and why Gemini 3 Flash is a killer implementation model. - ["QA is the Killer App for PI & Claude Coding Agents"](https://jprokay.com/post/019-qa-agents-not-coding-agents): Using persona-driven agents to automate end-to-end browser testing. - ["pi: The Coding Agent For Your Workflow"](https://jprokay.com/post/018-pi-coding-agent): A deep dive into the extensible agent harness that replaced Claude Code for my daily work. - ["Claude Code Custom Slash Commands: List & Workflow Guide"](https://jprokay.com/post/011-slash-commands): Practical examples and templates for orchestrating dev cycles with custom slash commands. ## All Posts - ["Claude Code Alternatives: Uncoupling from Anthropic with Pi Agent"](https://jprokay.com/post/020-uncoupling-from-anthropic): Learn about Claude Code alternatives and how to uncouple from Anthropic using the Pi Agent. Opus is not the only model worth your time—discover why Gemini 3 Flash and others might be the right fit. - ["QA is the Killer App for PI & Claude Coding Agents"](https://jprokay.com/post/019-qa-agents-not-coding-agents): Why QA is the best use case for pi agents and Claude Code. Learn how to use persona-driven agents and cmux for automated browser testing. - ["pi: The Coding Agent For Your Workflow"](https://jprokay.com/post/018-pi-coding-agent): Most coding agents (and especially their interfaces) have limited extensibility. pi is a minimal terminal harness you extend with TypeScript — custom tools, commands, and even the TUI. Here's how I built three extensions that turned it into an agentic IDE. - [Working on the Go with Terminus, Tailscale, and Claude](https://jprokay.com/post/017-mobile-dev-workflow): A look at using Terminus, Tailscale, and Claude Code to enable productive development sessions from anywhere. - [Looking back on a year of change](https://jprokay.com/post/016-reflections-on-2025): 2025 has been a transformative year both personally and professionally. Let's reflect and look forward to 2026 - [devenv - The best tool for the job](https://jprokay.com/post/015-devenv): You start a new project, you open up the README, and begin going through all the installation instructions. Oops someone forgot to document a dependency! What if you could get rid of this problem entirely? What if setting up a project just required you to open it up? - [Process update - How I use Agents as of Dec. 2025](https://jprokay.com/post/014-process-update): My process has solidified over the fall into something consistent and repeatable. Even as models change, I have found a lightweight, easy to manage flow with Claude. - [Finally Understanding Subagents](https://jprokay.com/post/013-finally-understanding-subagents): Creating subagent configurations for specific tasks in Claude Code or Opencode. - [Planning with Claude Desktop, Coding with Claude Code ](https://jprokay.com/post/012-claude-and-claude-code): Using different types of Claude to play to their strengths - ["Claude Code Custom Slash Commands: List & Workflow Guide"](https://jprokay.com/post/011-slash-commands): Explore a list of custom Claude Code slash commands to automate your dev workflow, from GitHub project setup to automated PRs. - [Remote Agents and You ](https://jprokay.com/post/010-remote-agents): Remote coding agents like Devin are getting a lot of hype. Are they ready for primetime? - [Helpful MCP Servers ](https://jprokay.com/post/009-mcp-suggestions): 3 MCP Servers that have made a difference - [Enterprise Vibe Coding ](https://jprokay.com/post/008-enterprise-vibes): Ideas on avoiding slop when you are trying to vibe - [Reformed AI Skeptic](https://jprokay.com/post/007-on-ai-assistance): AI assistants are changing coding. Is it still worth the skeptism? - [Shifting Left ](https://jprokay.com/post/006-shifting-left): AI is shifting the task of engineering towards one of judgement and evaluation. What skills will the engineer of the future need to be successful? - [Minimal Viable Blogging ](https://jprokay.com/post/005-mvb): Starting a blog doesn't require Rails, Django, or Astro. You can do it with HTML, CSS, and a bit of Caddy. Start your own Minimal Viable Blog - [Sorting files in reverse order ](https://jprokay.com/post/004-sorting-files): Using Sprig and Caddy template functions to render a list of posts in reverse order - [Styling code blocks with Chroma ](https://jprokay.com/post/003-styling-code-with-chroma): Setting up CSS with Chroma to render code blocks - [Anatomy of a Post ](https://jprokay.com/post/002-anatomy-of-a-post): Leveraging Caddy's splitFrontMatter to build a feed - [Hello World ](https://jprokay.com/post/001-first-post): Using Caddy templates to generate a link feed ala Simon Willison ## Technical Interests - AI-assisted development workflows - Model Context Protocol (MCP) and AI tool integration - Minimalist web architectures - Developer productivity and automation - Software engineering best practices ## Contact - [LinkedIn](https://www.linkedin.com/in/julian-prokay/) - [GitHub](https://github.com/jprokay) - [Bandcamp](https://bandcamp.com/jbeard) ## Machine-readable resources - [Agent overview](https://jprokay.com/agent.md) - [Profile, work history, links, and consulting](https://jprokay.com/api/profile.json) - [Post metadata](https://jprokay.com/api/posts.json) - [API documentation](https://jprokay.com/docs/api.md) - [OpenAPI description](https://jprokay.com/openapi.json) - [Agent skill index](https://jprokay.com/.well-known/agent-skills/index.json) ## Optional - [Homepage](https://jprokay.com/): Personal website and blog - [Blog Feed](https://jprokay.com/feed): Complete list of all blog posts
Share or continue the analysis
Share this audit
Discuss with AI
Added 8/3/2026