Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Agent Skills

This repository is my personal collection of agent skills. Each skill packages focused instructions and any supporting resources an agent needs to carry out a repeatable workflow.

ralphex-plan-writer

ralphex-plan-writer creates verification-first Markdown implementation plans compatible with Ralphex and executr. An agent should trigger it when asked to turn a PRD or brainstorming note into autonomous implementation tasks, design an executr verification loop, or prepare a Ralphex-compatible plan. It emphasizes minimal, approval-gated architecture, deterministic verification, and validation with its bundled format checker. Install it with npx skills add github.com/luisKisters/skills --skill ralphex-plan-writer.

hyperkey-setup

hyperkey-setup installs or configures Karabiner-Elements with a bundled Caps Lock Hyperkey and right Shift launcher mapping. An agent should trigger it when a user asks for this keyboard setup, wants Caps Lock to act as Hyper when held and Cmd+Tab when tapped, or wants right Shift to launch Cmd+Space or Raycast when tapped. It handles installation checks, safely copies the predefined rules, guides or automates setup where possible, and verifies both shortcuts. Install it with npx skills add github.com/luisKisters/skills --skill hyperkey-setup.

stop-yapping

stop-yapping stops rambling while keeping responses short, clear, self-contained, and complete. An agent should trigger it when a user says the agent is yapping, rambling, too verbose, or unclear, or asks for shorter, direct, ADHD-friendly communication. If invoked alone, it answers the user's previous substantive message with these rules. Install it with npx skills add github.com/luisKisters/skills --skill stop-yapping.

orchestrated-build

orchestrated-build executes a written implementation plan with a fleet of agents: one Opus 5 low subagent per vertical slice, each supervising a codex exec gpt-5.6-sol run in tmux, with 10-minute check-ins, one review phase, and one end-to-end phase at the end. An agent should trigger it when a user asks to execute or ship a plan, delegate slices to subagents, parallelize a build, or supervise long autonomous runs. Install it with npx skills add github.com/luisKisters/skills --skill orchestrated-build.

How to install a skill

Install one skill with npx skills add <repo> --skill <name>. For this repository, replace <repo> with github.com/luisKisters/skills and <name> with the skill name shown above.

If the CLI is unavailable, clone or download the repository and manually copy the selected skill directory into your agent's skills directory, for example cp -R <skill-directory> ~/.codex/skills/<name>.

About

Skills I came up with:)

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages