Designing an Agent Reading Test
In which I try to give people tools to understand how agents read web content, and where they fail.
21 Posts
Ai
Measure Agent Web Traffic Redux
In which I revisit measuring agent web traffic, and dive deeper down the rabbit hole.
Confident-Sounding Gibberish
In which I share research about why LLM-generated output is hard to fact check.
The Verification Gap in AI Content Pipelines
In which verification is the hardest unsolved problem in AI content pipelines, and most organizations don't know it.
How to Evaluate a Platform-Written Spec
In which I find that a platform-published spec's omissions track financial incentives.
Drafting Editorial Content with AI
In which I use AI to help draft content, and discover its limitations.
Filtering AI News
In which I build a system to filter AI-related content.
When a Feature Request Becomes a Research Project
In which someone asks for something in 'skill-validator' and I spin up a new community research project.
Why a Platform Shouldn't Own an Open Spec
In which a platform breaks the Agent Skills spec for its own benefit.
Is Your llms.txt Already Stale?
In which I build a freshness check for llms.txt and discover that my tools were the problem.
Agent Skill Mega Repo Woes
In which I validate a 23.7k-star skill mega repo and discover problems the star count won't tell you.
Vibes are Out, Data is In
In which I explain why a vibes-based approach to AI and docs ain't cutting it.
How to Measure Agent Web Traffic
In which I lurk in server logs and try to make sense of what I see.
An Agent is More Than Its Brain
In which I talk about what's inside a coding agent.
Make Your Hugo Site Agent-Friendly
In which I show you how I made multiple Hugo sites agent-friendly.
Case Study - 'upgrade-stripe' Agent Skill
In which I deep dive on a Stripe Skill, and what it means for the industry.
LLMs vs. Agents as Docs Consumers
In which I discuss how model training and agents need different docs access patterns.
Upskilling in the AI Age
In which I answer someone who asked me how to get started with AI.
Agent Web Fetch Spelunking
In which I ask an agent to help me explore Web Fetch with docs pages - and get a surprise.
Agent-Friendly Docs
In which I ask an agent to view hundreds of docs pages - and feel sad.
Agent Skill Analysis
In which I build a validator and analyze popular Agent Skills.