Scaffold a Ralph Loop — a self-driving bash loop that repeatedly invokes Claude to implement one task at a time until the work is complete. Drives either a single GitHub issue or a whole PRD's linked child issues in sequence. Lists open ready-for-agent GitHub issues, lets the user pick one, and creates a ralph-it/ folder at the repo root with a hardened ralph.sh tailored to that target plus a README. Use when the user wants to set up a Ralph loop, autonomously grind through a PRD or issue, "ralph it", or run an unattended implement-test-commit loop.
Apply this skill ONLY when the user explicitly asks for it. Trigger on direct requests such as "use human voice", "apply the human-voice skill", "make it sound human", "remove the AI feel", "less robotic", "don't sound like ChatGPT", "write naturally", "avoid AI cliches", or "humanize this text". Do NOT trigger proactively on general writing, rewriting, or editing tasks (emails, posts, reports, docs) unless the user explicitly requests human-sounding output in their message.
A personal collection of Claude Code skills, packaged as an installable plugin. Each skill teaches Claude how to perform a specific workflow; they activate automatically when your request matches a skill's trigger, or you can invoke one explicitly.
Add this repo as a plugin from within Claude Code:
/plugin install dcampillo/skills
Once installed, the skills below become available. Claude loads each one on demand based on what you ask for.
| Skill | What it does | Use when |
|---|---|---|
ralph-it | Scaffolds a Ralph Loop — a self-driving bash loop that repeatedly invokes Claude to implement one task at a time until the work is done. Drives a single GitHub issue or a whole PRD's linked child issues in sequence, with a hardened ralph.sh, completion gates, and a no-progress guard. | You want to set up an unattended implement → test → commit loop against a GitHub issue or PRD ("ralph it", "grind through this PRD"). |
| Skill | What it does | Use when |
|---|---|---|
human-voice | Helps you write with a natural, authentic human voice that feels genuine and conversational while maintaining clarity and impact. | You want to improve the tone and authenticity of your writing, make it more relatable, or ensure your voice comes through clearly in prose, emails, or documentation. |
| Skill | What it does | Use when |
|---|---|---|
topical-authority-map | Builds a pillar-and-cluster topical authority map for a niche, exported as a CSV with titles, slugs, search intent, estimated volume, and keyword difficulty. | You want to plan a site's content around a niche, build topical authority, or need a keyword/content map ("topical map", "content plan", "what should I write to rank"). |
.
├── .claude-plugin/
│ └── plugin.json # plugin manifest (lists each skill)
├── skills/
│ └── <category>/
│ └── <skill-name>/
│ └── SKILL.md # the skill (plus any bundled resources)
├── CLAUDE.md # contributor guide for adding skills
├── LICENSE
└── README.md
See CLAUDE.md for the conventions on adding a new skill to this repo.
MIT © David Campillo
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub dcampillo/skills --plugin dcampillo-skillsPlugin-safe Claude Code distribution of Antigravity Awesome Skills with 1,561 supported skills.
Unity Development Toolkit - Expert agents for scripting/refactoring/optimization, script templates, and Agent Skills for Unity C# development
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Persistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).