Effective HTML
Skillplannotator
Skills for turning agent output into useful HTML artifacts: wireframes, interactive prototypes, plans and diagrams.
Skills you drop into Claude Code, Claude Desktop, Codex, or Cursor to teach it a job. Sort by trending, rating, or upvotes.
1394 skills
plannotator
Skills for turning agent output into useful HTML artifacts: wireframes, interactive prototypes, plans and diagrams.
steipete
Peter Steinberger's shared agent skills and scripts, from browser use and transcripts to domain registration, reused across his repos.
samber
About 46 Go skills for coding agents covering concurrency, context, benchmarks, CLI design, code style and CI.
htdt
Let Claude Code or Codex build Godot games end to end, including generating the game's art assets.
ParthJadhav
Build exportable App Store and Google Play marketing screenshots from a Next.js screenshot editor your agent scaffolds.
Sapient Intelligence
Gather literature, benchmarks and prior art for an autonomous research task, with provenance notes.
Amos Blomqvist
A teaching skill that aims for real understanding by building each idea from foundations you already know.
edenfunf
Show it a video you like and an agent crew builds a new one in the same style.
Jake Schincariol
Eleven skills that study an app, rebuild its features clean-room style, test it and launch it.
AminBlg
Make your agent write docs like a maintenance manual: short, plain and defined.
danyuchn
Rewrite tool descriptions, prompts and errors in Simplified Technical English so agents can't misread them.
HKUST DIAL
Twelve research-advisor skills, from evaluating an idea to drafting a rebuttal.
Tigerless Labs
Shared long-term memory for Claude Code and Codex, stored as plain Markdown with local retrieval.
Nate Herk
Build premium scroll-driven landing pages with layered heroes and verified mobile states.
Jared Palmer
Strip AI-generated slop from a branch diff without changing behavior.
Jared Palmer
An extremely strict structural review of a branch: abstractions, duplication, sprawl and files past 1k lines.
NVIDIA Research
Hand a Claude Code prompt to NVIDIA's SoL-Pi harness, with the session passed along as context.
Leonxlnx
Stop agents from quitting early: write acceptance gates first, then prove the work is done.
Inkeep
Proposals, decisions, specs, design reviews and postmortems as agent workflows in a markdown wiki.
Vercel Labs
Write and edit SDK docs, guides and API references in Vercel's editorial voice.
Anthropic
The guardrails Anthropic's commerce agents enforce for untrusted content, writes, identity and memory.
Anthropic
Write and run behavioral evals for shopping and merchant agents, poisoned fixtures included.
Anthropic
Anthropic's reference shopping agent and merchant agent, each with its own skill set.
Google DeepMind
Score and annotate genetic variants with AlphaGenome Variant Impact scores via the Atlas API.