#ai

An Agent Skill license can do more than say who may use what: it can teach agents how to comply, meter use, and leave a verifiable trail — without hidden telemetry and without turning licensing into a billing framework.

A literary system where you earn the right to judge the canon by extending it — and what happens when the blockchain is a DuckDB file in git.

A pairwise ranking system for blog posts — and what happened when an AI agent filled required reviews with random tokens instead of actual criticism.

Indexing the AI economy looks hard until you stop thinking like a buyer and start thinking like a tax collector.

On three professional postures, four alignment properties, and the one property that had to come from elsewhere.

On writing the paper before doing the research, and other engineering practices that should embarrass us less than they do.

On Cucumber, content-addressing, and an alignment technique that turns out to be older than alignment.

On Tinkerbell, persona prompts, and why declaring the frame is what kills the play.

How a single word has been quietly summoning Waluigis for half a decade, and what the swiss-army knife in my coat pocket has to do with it.

My father records stories on his phone. Jules committed the wrong year. This is what I built from that.

The sandbox separates draft from act. What it doesn't do is answer where the accountability lives when the sandbox fails.

Using Jules to generate a novel about a father speaking with his children from the future — and why it echoes Kleber Mendonça Filho's O Agente Secreto.

The identity-repo pattern: an agent's memory and its cognitive engine are separable. What this looks like in practice — and why it matters more than it sounds.