Latest

Browse and install thousands of AI Agent skills in the Killer-Skills directory. Supports Claude Code, Windsurf, Cursor, and more.

This directory brings installable AI Agent skills into one place so you can filter by search, category, topic, and official source, then install them directly into Claude Code, Cursor, Windsurf, and other supported environments.

3407 available skills

openspec-main-spec-format

Logo of timurb
timurb

Install CLI npm utils as docker container for increased security

0
0
Developer

sam

Logo of OpenPeopleStudio
OpenPeopleStudio

OpenPeople.ai - AI-powered commerce infrastructure for ambitious retail brands

0
0
Developer

git-commit-formatter

Logo of uta8a
uta8a

information gathering tool

0
0
Developer

openai-gpt5-prompting

Logo of ijindal1
ijindal1

jaunt, dont code.

0
0
Developer

plan-confirm-workflow

Logo of Birdiz
Birdiz

Config files for my GitHub profile.

0
0
Developer

security-audit

Logo of duhenri9
duhenri9

Landing page oficial da WM3, criada para consolidar o reposicionamento da agência no mercado digital. Centraliza todos os links, serviços e projetos em um só hub estratégico, refletindo a nova fase da marca e a integração de soluções de automação e IA.

0
0
Developer

inspequte-rule-spec

Logo of KengoTODA
KengoTODA

Fast, CLI-first static analysis tool for JVM class and JAR files. Designed for coding agents.

0
0
Developer

checkpoint-ambiguity-review

Logo of SprocketLab
SprocketLab

SlopCodeBench: Measuring Code Erosion Under Iterative Specification Refinement

0
0
Developer

explore

Logo of sylin-org
sylin-org

Local mDNS service discovery for everyone.

0
0
Developer

log-debug

Logo of jihyukma123
jihyukma123

repository to test whether skills works on codex sdk

0
0
Developer

lcp-protocol-spec

Logo of YusukeShimizu
YusukeShimizu

LCP (Lightning Compute Protocol) is an application-layer protocol for paying for small compute jobs over Lightning.

0
0
Developer

opener-variator

Logo of WILLOSCAR
WILLOSCAR

Research pipelines as semantic execution units: each skill declares inputs/outputs, acceptance criteria, and guardrails. Evidence-first methodology prevents hollow writing through structured intermediate artifacts.

0
0
Developer