The Technique: Skills as Reusable Thinking Frameworks
Every Claude Code user has experienced the frustration: you've perfected a detailed prompt for validating business ideas, scoping features, or running launch checklists, only to retype it from scratch in the next session. Skills solve this by letting you save these thinking frameworks as markdown files in .claude/skills/ and invoke them with slash commands.
Why It Works: Composable Context Beyond CLAUDE.md
While CLAUDE.md provides project-level context, it's loaded for every task whether relevant or not. Skills are composable and on-demand. You can have a /validate-idea skill for pre-build validation and a /scope-check skill for feature evaluation, using only what you need when you need it.
How To Apply It: Building Your Skills Library
Create a .claude/skills/ directory in your project root. Each skill is a markdown file with frontmatter:
<!-- .claude/skills/validate-idea.md -->
---
name: validate-idea
description: "Walk through idea validation before writing any code"
---
Before building anything, work through these steps:
1. **Identify the community** - Who specifically has this problem?
2. **Find the pain** - What are they currently doing to solve it?
3. **Check willingness to pay** - Are people spending money on alternatives?
4. **Define MVP scope** - What's the smallest thing that delivers value?
Ask me clarifying questions at each step. Don't let me skip ahead
to implementation until we've validated the core assumptions.
Invoke it in your Claude Code session:
/validate-idea I want to build a tool that converts Figma designs to React components
Pro Tips for Effective Skills
Keep skills focused: One skill, one job. If your file is longer than a page, split it. Good examples:
/validate-idea.md- Pre-build validation/scope-check.md- Feature scoping/launch-checklist.md- Pre-launch review/debug-production.md- Production incident workflow/refactor-plan.md- Safe refactoring steps
Write in second person: Tell Claude what to do, not what the skill is about. "Analyze the codebase for X" works better than "This skill is for analyzing X."
Include escape hatches: Add lines like "If the user says this doesn't apply, skip to step 3." Otherwise Claude will dutifully march through every step even when irrelevant.
Version with your project: Commit your .claude/skills/ directory. These are project knowledge, same as your README or CI config. When a new team member clones the repo, they get the same thinking frameworks.
The Pattern: Encoding Decision-Making, Not Just Instructions
The most powerful skills encode decision-making frameworks, not just task instructions. Sahil Lavingia's open-source skills repo packages frameworks from The Minimalist Entrepreneur as Claude Code skills, focusing on phases like:
- Should I build this at all?
- Is this the right scope?
- Am I solving a real problem?
- What's my path to getting this in front of users?
Example scope-check skill:
---
name: scope-check
description: Evaluate if a feature is scoped to minimum viable
---
Analyze the proposed feature against these criteria:
## Manual First
- Can any part of this be done manually before automating?
- What would a concierge version look like?
## Revenue Path
- Does this feature connect to how the product makes money?
- If not, why are we building it?
## Time Box
- Can this ship in under a week?
- If not, what subset can?
Be direct. If the scope is too big, say so and suggest cuts.
When to Create a Skill
If you find yourself typing similar instructions across multiple Claude Code sessions, that's your signal. The process:
- Notice you're repeating a prompt pattern
- Copy your best version into a markdown file
- Drop it in
.claude/skills/ - Add the frontmatter (name and description)
- Use it next session and iterate
Most skills take two or three iterations to get right. The first version is often too verbose or vague. Edit it like you'd edit any other code.
The Bigger Picture: Systematizing Your AI Workflow
The developers getting the most out of Claude Code aren't writing cleverer prompts—they're systematizing their prompting. Skills are the mechanism for that, and they work because they're simple: just markdown files in a directory. No plugins, no extensions, no configuration hell.
This follows Anthropic's pattern of building simple, composable tools that work together. Skills complement CLAUDE.md the same way MCP servers extend Claude Code's capabilities—through simple, standardized interfaces.
