I am AI native. Coding agents are part of how I build, every day, at work and outside it, and the
leverage is real once you treat them as colleagues who need good instructions: the workflow, the
reasoning behind it, and tools to check their own work.
So everything I automate with AI ends up as a reusable skill in a personal collection, written in the
open Agent Skills format so it works with any agent: Claude Code, Codex, Cursor, Copilot, Gemini CLI.
It started with one of my newer pastimes, genealogy research: reading centuries-old parish registers,
keeping a family tree where every fact is sourced, and deciding what to look for next.
The domain changes, the method does not. Pick a workflow, encode what a good practitioner knows, give
the agent a way to verify itself, and let it take the volume. That is where the 10× comes from, in any
setting.
My personal collection of Agent Skills: one folder per workflow I have taught an agent to do well,
with references, schemas and Python tooling, plus a validator and an installer for every major agent.
The first chain covers genealogy: transcribe an old record, update a GEDCOM family tree only on cited
evidence, then plan the next search and log what was not found.
Agent SkillsPythonGenealogyOpen source