#283283rd of 389 in Skills, plugins and prompts
skills
Claude Code skills for research-to-implementation, PR outlines and CLAUDE.md cleanup
- Stars
- 5.7k
- License
- MIT
- Last commit
- Nov 2026
- Language
- TypeScript
83 on 20 Apr 20265,731 on 9 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
A collection of installable Claude Code skills from HumanLayer, added one at a time with `npx skills add humanlayer/skills --skill NAME`. Skills include rpi (research, design discussion and implementation from local documents), visual-pr, show-me, improve-claude-md, narrow-react-prop-types, and two that set up scheduled coding-agent loops in GitHub Actions.
Who it is for: Claude Code users wanting workflow skills for planning, PRs and agent loops
Strengths
- Each skill installs separately via one npx command
- rpi offers four workflows and runs each phase in a fresh session by default
- Includes skills that generate GitHub Actions workflows for scheduled coding-agent runs
- Also works in Codex using `$rpi` instead of `/rpi`
Weaknesses
- Codex's default sandbox can make `.agents` read-only, so rpi needs write access granted
- Some skills are narrow, such as narrow-react-prop-types for React only
- No releases published; versioning relies on the repo head
- Skill descriptions are brief, with no documented examples of output
What it needs
- no GPU
- Needs Node.js (npx), Claude Code, GitHub Actions (for loop skills)
- Models: Claude Code, Codex
Quick answers
Is skills open source?
Yes. Its code is on GitHub under the MIT license.
Also in Skills, plugins and prompts
See all 389| Rank | Project | Score |
|---|---|---|
| 1 | claude-memPersistent cross-session memory plugin for Claude Code and other coding agents | 97 out of 100 |
| 2 | impeccableDesign skill with 24 slash commands for AI coding agents' frontend output | 97 out of 100 |
| 3 | gstackMarkdown slash-command skills that give Claude Code planning, review, QA and ship roles | 96 out of 100 |
| #4 | rufloAgent harness adding swarms, memory and hooks to Claude Code and Codex | 96 out of 100 |
| #5 | cavemanShortens coding-agent replies and compresses tool output before the model reads it | 95 out of 100 |