Skills

Agent Skills I Actually Run

Install-ready skills for Claude Code, Codex, and any coding agent — the same ones I use to operate five AI-native companies. Free. Drop the folder in, run the command.

Planning /roadmap

The Roadmap Skill

Run your roadmap as a markdown file your agents read and update — no Linear, no Jira.

The exact skill I use to run the roadmap across the five companies I operate. It turns a single roadmap.md (plus per-item files) into the source of truth your coding agents read, rank, and update — four modes: bootstrap a repo's scattered TODOs into one file, pick what to work on next, mutate items as state changes, and scrub the whole thing clean. I deleted Linear; this is what replaced it.

Claude Code · Codex · any agent

Get the skill

Engineering /agent-harness

Agent Harness

Audit and optimize your agent harness — the scaffolding that makes AI coding agents effective in your repo.

The skill I run on a repo before I trust agents to ship in it. It scores your harness — CLAUDE.md / AGENTS.md structure, documentation routing, definition of done, automated verification, knowledge persistence, task hygiene — against a rubric, shows you the scorecard, then fixes what you approve. This is the difference between agents that flail and agents that operate.

Claude Code · Codex · any agent

Get the skill

How it works

Pick a skill

Each one is a real skill I run, not a demo. Folder of SKILL.md + references.

Enter your email

I send the .zip straight to your inbox — no payment, no waiting.

Drop it in & run

Unzip into your agent's skills directory, start a session, run the command. Done.