A growing collection of custom Claude skills — SaaS engineering, debugging, AI prompting, content scripting, and more. Continuously expanding.
SaferSkills independently audited Claude-Skills (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
A growing collection of Claude skills that unlock expert-level performance across engineering, debugging, AI tools, content creation, and more!
Globally applicable. Any stack. Any language. Continuously expanding.
Claude Skills are instruction files that Claude reads before tackling specific tasks. Instead of generic AI responses, Claude behaves like a domain expert — following battle-tested patterns, using the right tools, and producing production-grade output.
Drop any SKILL.md into a Claude Project's knowledge base and it activates automatically for every relevant conversation in that project.
| Skill | What It Does | Triggers On |
|---|---|---|
saas-builder | Full-stack SaaS engineering — any language, stack, or market | Product features, startup architecture, API design |
debug-fast | Root-cause debugging in 12+ languages with deep-diagnosis tools | Any bug, error, crash, or "this isn't working" |
prompt-engineer | Optimal prompts for every major AI — image, video, audio, LLMs, code | "Write a prompt for...", "How do I get better results from X" |
content-pipeline | Scripts for 7 content formats across any topic | "Write a script for...", "Plan a video about..." |
memory-vault | Retrieve anything from past conversations — code, decisions, designs | "Find the code you wrote for...", "What did we decide about..." |
More skills being added. Star this repo to get notified.
Skills being planned or in progress:
| Skill | Description |
|---|---|
code-reviewer | Deep code review across any language — security, performance, style |
data-analyst | SQL, pandas, data visualization, and insight generation |
system-design | Architecture diagrams, scalability patterns, tradeoff analysis |
marketing-copy | Landing pages, ads, email sequences, product descriptions |
research-agent | Structured web research, source evaluation, report generation |
Have an idea for a skill? Open an issue or submit a PR.
Claude-Skills/
├── README.md
├── saas-builder/
│ ├── SKILL.md
│ └── references/
│ ├── payment-providers.md
│ ├── auth-patterns.md
│ ├── db-patterns.md
│ └── api-patterns.md
├── debug-fast/
│ └── SKILL.md
├── prompt-engineer/
│ └── SKILL.md
├── content-pipeline/
│ └── SKILL.md
└── memory-vault/
└── SKILL.mdEach skill lives in its own folder. Large reference material goes in a references/ subfolder to keep the main SKILL.md focused and readable.
SKILL.md file(s) you want activecat */SKILL.md > combined-skills.md
# Upload combined-skills.md as project knowledgePaste the contents of any SKILL.md into your system prompt or CLAUDE.md.
Skills are plain Markdown — easy to write, easy to improve.
To add a new skill:
your-skill-name/SKILL.md with YAML frontmatter:---
name: your-skill-name
description: >
One paragraph describing what this skill does and exactly when Claude
should trigger it. Be specific and aggressive with trigger phrases.
---your-skill-name/references/SKILL.md under 500 linesTo improve an existing skill:
critical, high, medium, lowMIT — use freely, fork, adapt, share.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.