Skip to content
Conduit

skill

claude-skill-pack

Claude skills organized into domain layers — engineering, product, founder/GTM, design, and documents

BTN101/claude-skill-packcommunitylistedunreviewed

Set up

Skills are installed by placing their folder where your agent loads skills.

The download comes from github.com/BTN101/claude-skill-pack itself, not from our server, so what you get is exactly what its author published.

Install instructions come from the README

Skills do not use an MCP config. Follow the steps in the README below.

README

synced Oct 23, 2026

claude-skill-pack

A curated collection of Claude skills organized into domain layers — covering engineering, product thinking, founder/GTM work, design quality, and document creation.

Built for use with Claude's skill system on claude.ai.


What are Claude Skills?

Skills are structured instruction sets that tell Claude how to approach a specific task — what workflow to follow, what quality gates to apply, what patterns to enforce. They're loaded into context when a matching task comes up, giving Claude specialist-level discipline rather than generic help.

Each skill lives in a folder with a SKILL.md file. The frontmatter description field is what triggers the skill — Claude reads it and decides whether to apply the skill based on your request.

skill-name/
├── SKILL.md          # Required: name, description, instructions
└── references/       # Optional: supporting docs loaded on demand

Skill Layers

This pack introduces layer skills — meta-skills that group related specialists together. Instead of needing to know which exact skill to reach for, you describe the task and the layer routes to the right one automatically.

LayerDomainSkills Inside
founder-layerBusiness, investors, GTM, contentinvestor-materials, investor-outreach, market-research, brand-voice, copywriting, article-writing, launch-strategy, pricing-strategy, competitive-landscape
engineering-layerCode, architecture, APIs, securityapi-design, backend-patterns, security-review, react-patterns, react-performance, nextjs-app-router-patterns, postgres-patterns, database-migrations, deployment-patterns, tdd-workflow, java-coding-standards
security-layerSecurity auditing, vulnerability hunting, vibe-coded app hardeningsecurity-review, find-bugs, gha-security-review, code-review
product-layerProduct strategy, discovery, specsproduct-lens, lean-startup, 37signals-way, mom-test, jobs-to-be-done, prioritization-advisor, prd-development, hooked-ux, user-story
design-layerUI quality, motion, anti-slopimpeccable, design-taste-frontend, redesign-existing-projects, framer-motion-animator, animation-micro-interaction-pack
document-layerFile creation and readingdocx, pptx, xlsx, pdf, file-reading, pdf-reading

How It Works

You:    "Cold email an investor about my fraud detection startup"
         ↓
Layer:  founder-layer triggers on "investor" + "email"
         ↓
Skill:  investor-outreach runs its structured workflow
         ↓
Output: A tight, personalized, proof-first email — not generic copy

The layer skill is the router. The individual skills are the knowledge. Both need to be installed.


Skills in This Pack

🟠 Layers (skills/layers/)

SkillWhat it does
founder-layerRoutes business, investor, GTM, and content tasks
engineering-layerRoutes code, architecture, security, and deployment tasks
security-layerRoutes security audits, vulnerability hunts, and vibe-coded app hardening
product-layerRoutes product strategy, discovery, and spec tasks
design-layerRoutes UI quality, motion, and design audit tasks
document-layerRoutes file creation and reading tasks

⚙️ Engineering (skills/engineering/)

SkillWhat it does
security-reviewSecrets management, input validation, SQL injection, XSS, CSRF, auth, RLS
find-bugsBranch-diff scanner — maps attack surface, runs OWASP checklist per changed file
gha-security-reviewGitHub Actions audit — pwn-request, expression injection, supply chain, credential theft
code-reviewHolistic PR review — runtime errors, N+1s, perf, backwards compat, security
api-designREST naming, status codes, pagination, versioning, error format
backend-patternsRepository + service layer, caching, background jobs, middleware
react-patternsComposition, compound components, custom hooks, context
react-performanceMemoization, code splitting, Server Components, waterfall fixes
react-testingTesting Library, Vitest/Jest, MSW mocking, accessibility assertions
postgres-patternsQuery optimization, indexing, connection pooling, RLS
database-migrationsSafe schema changes, rollbacks, zero-downtime migrations
deployment-patternsVercel, Docker, CI/CD, health checks, rollback strategy
tdd-workflowTest-driven development, 80%+ coverage target
java-coding-standardsSpring Boot / Quarkus naming, immutability, streams, exceptions

🧠 Founder (skills/founder/)

SkillWhat it does
investor-materialsPitch decks, one-pagers, financial models, accelerator applications
investor-outreachCold emails, warm intros, follow-ups — proof over adjectives
market-researchTAM/SAM/SOM sizing, competitive analysis, investor diligence
brand-voiceReusable voice profile from real source material
article-writingBlog posts, articles, technical essays, newsletters

🔭 Product (skills/product/)

SkillWhat it does
product-lensFounder diagnostic — who is this for, what's the pain, what's the MVP

Installation

Install a .skill file on claude.ai

  1. Download the .skill file from the releases page
  2. Go to claude.ai → Settings → Skills
  3. Click Install skill and select the .skill file
  4. The skill is now active across your conversations

Install from source

Clone the repo and use any SKILL.md directly — the file format is plain markdown with YAML frontmatter.

git clone https://github.com/BTN101/claude-skill-pack

Skill File Format

Every skill follows this structure:

---
name: skill-name
description: When to trigger and what the skill does. This is the
             primary mechanism — Claude matches your request against
             all skill descriptions to decide what to load.
---

# Skill Name

Instructions for Claude go here in plain markdown.

## When to Activate
## Workflow
## Quality Gate

The description field is the most important part. It acts as the trigger — so it needs to be specific about both what the skill does and when it should fire.


Contributing

Skills are just markdown files. To add one:

  1. Create a folder under the relevant category in skills/
  2. Add a SKILL.md with proper YAML frontmatter
  3. Optionally add a references/ folder for supporting docs
  4. Submit a PR

Good skills have: a precise trigger description, a clear workflow, and a quality gate before output.


Why This Exists

Most AI assistance is generic. Skills make it structural — the same way a senior engineer follows a checklist before shipping an API endpoint, or a founder applies a framework before deciding what to build. This pack is an attempt to encode that discipline into reusable, shareable form.


License

MIT — use freely, modify as needed, attribution appreciated.


Built by @BTN101 · Centurion, South Africa

↑↓ navigate openesc close