Writing Skills

Overview

Writing skills IS Test-Driven Development applied to process documentation.

Personal skills live in agent-specific directories (~/.claude/skills for Claude Code)

You identify common failure patterns (baseline behavior), write the skill (documentation) addressing those patterns, then verify through application scenarios, and refactor (close loopholes).

Core principle: If you didn't identify what agents naturally do wrong, you don't know if the skill prevents the right failures.

REQUIRED BACKGROUND: You MUST understand test-driven-development before using this skill. That skill defines the fundamental RED-GREEN-REFACTOR cycle. This skill adapts TDD to documentation.

Official guidance: For Anthropic's official skill authoring best practices, see anthropic-best-practices.md. This document provides additional patterns and guidelines that complement the TDD-focused approach in this skill.

Complete worked example: See examples/CLAUDE_MD_TESTING.md for a full verification campaign testing CLAUDE.md documentation variants.

What is a Skill?

A skill is a reference guide for proven techniques, patterns, or tools. Skills help future Claude instances find and apply effective approaches.

Skills are: Reusable techniques, patterns, tools, reference guides

Skills are NOT: Narratives about how you solved a problem once

TDD Mapping for Skills

The entire skill creation process follows RED-GREEN-REFACTOR.

When to Create a Skill

Create when:

• Technique wasn't intuitively obvious to you
• You'd reference this again across projects
• Pattern applies broadly (not project-specific)
• Others would benefit

Don't create for:

• One-off solutions
• Standard practices well-documented elsewhere
• Project-specific conventions (put in CLAUDE.md)

Skill Types

Technique

Concrete method with steps to follow (condition-based-waiting, root-cause-tracing)

Pattern

Way of thinking about problems (flatten-with-flags, test-invariants)

Reference

API docs, syntax guides, tool documentation (office docs)

Directory Structure

skills/
  skill-name/
    SKILL.md              # Main reference (required)
    supporting-file.*     # Only if needed

Flat namespace - all skills in one searchable namespace

Separate files for:

1. Heavy reference (100+ lines) - API docs, comprehensive syntax
2. Reusable tools - Scripts, utilities, templates

Keep inline:

• Principles and concepts
• Code patterns (< 50 lines)
• Everything else

SKILL.md Structure

Frontmatter (YAML):

• Only two fields supported: name and description
• Max 1024 characters total
• name: Use letters, numbers, and hyphens only (no parentheses, special chars)
• description: Third-person, includes BOTH what it does AND when to use it
  • Start with "Use when..." to focus on triggering conditions
  • Include specific symptoms, situations, and contexts
  • Keep under 500 characters if possible

---
name: Skill-Name-With-Hyphens
description: Use when [specific triggering conditions and symptoms] - [what the skill does and how it helps, written in third person]
---

# Skill Name

## Overview

What is this? Core principle in 1-2 sentences.

## When to Use

[Small inline flowchart IF decision non-obvious]

Bullet list with SYMPTOMS and use cases
When NOT to use

## Core Pattern (for techniques/patterns)

Before/after code comparison

## Quick Reference

Table or bullets for scanning common operations

## Implementation

Inline code for simple patterns
Link to file for heavy reference or reusable tools

## Common Mistakes

What goes wrong + fixes

## Real-World Impact (optional)

Concrete results

Claude Search Optimization (CSO)

Critical for discovery: Future Claude needs to FIND your skill

1. Rich Description Field

Purpose: Claude reads description to decide which skills to load for a given task. Make it answer: "Should I read this skill right now?"

Format: Start with "Use when..." to focus on triggering conditions, then explain what it does

Content:

• Use concrete triggers, symptoms, and situations that signal this skill applies
• Describe the problem (race conditions, inconsistent behavior) not language-specific symptoms (setTimeout, sleep)
• Keep triggers technology-agnostic unless the skill itself is technology-specific
• If skill is technology-specific, make that explicit in the trigger
• Write in third person (injected into system prompt)

# ❌ BAD: Too abstract, vague, doesn't include when to use
description: For async testing

# ❌ BAD: First person
description: I can help you with async tests when they're flaky

# ❌ BAD: Mentions technology but skill isn't specific to it
description: Use when tests use setTimeout/sleep and are flaky

# ✅ GOOD: Starts with "Use when", describes problem, then what it does
description: Use when tests have race conditions, timing dependencies, or pass/fail inconsistently - replaces arbitrary timeouts with condition polling for reliable async tests

# ✅ GOOD: Technology-specific skill with explicit trigger
description: Use when using React Router and handling authentication redirects - provides patterns for protected routes and auth state management

2. Keyword Coverage

Use words Claude would search for:

• Error messages: "Hook timed out", "ENOTEMPTY", "race condition"
• Symptoms: "flaky", "hanging", "zombie", "pollution"
• Synonyms: "timeout/hang/freeze", "cleanup/teardown/afterEach"
• Tools: Actual commands, library names, file types

3. Descriptive Naming

Use active voice, verb-first:

• ✅ creating-skills not skill-creation
• ✅ writing-skills not skill-writing

4. Token Efficiency (Critical)

Problem: getting-started and frequently-referenced skills load into EVERY conversation. Every token counts.

Target word counts:

• getting-started workflows: <150 words each
• Frequently-loaded skills: <200 words total
• Other skills: <500 words (still be concise)

Techniques:

Move details to tool help:

# ❌ BAD: Document all flags in SKILL.md
search-conversations supports --text, --both, --after DATE, --before DATE, --limit N

# ✅ GOOD: Reference --help
search-conversations supports multiple modes and filters. Run --help for details.

Use cross-references:

# ❌ BAD: Repeat workflow details

When implementing feature, follow these 20 steps...
[20 lines of repeated instructions from another skill]

# ✅ GOOD: Reference other skill

For implementation workflow, REQUIRED: Use [other-skill-name] for complete process.

Compress examples:

# ❌ BAD: Verbose example (42 words)

your human partner: "How did we handle authentication errors in React Router before?"
You: I'll search for React Router authentication patterns in our codebase and documentation to find previous implementations.
[Detailed explanation of search process...]

# ✅ GOOD: Minimal example (15 words)

Partner: "How did we handle auth errors in React Router?"
You: [Search codebase → provide solution]

Eliminate redundancy:

• Don't repeat what's in cross-referenced skills
• Don't explain what's obvious from command
• Don't include multiple examples of same pattern

Verification:

wc -w skills/path/SKILL.md
# getting-started workflows: aim for <150 each
# Other frequently-loaded: aim for <200 total

Name by what you DO or core insight:

• ✅ condition-based-waiting > async-test-helpers
• ✅ using-skills not skill-usage
• ✅ flatten-with-flags > data-structure-refactoring
• ✅ root-cause-tracing > debugging-techniques

Gerunds (-ing) work well for processes:

• creating-skills, testing-skills, debugging-with-logs
• Active, describes the action you're taking

4. Cross-Referencing Other Skills

When writing documentation that references other skills:

Use skill name only, with explicit requirement markers:

• ✅ Good: **REQUIRED SUB-SKILL:** Use superpowers test-driven-development
• ✅ Good: **REQUIRED BACKGROUND:** You MUST understand superpowers systematic-debugging
• ❌ Bad: See skills/testing/test-driven-development (unclear if required)
• ❌ Bad: @skills/testing/test-driven-development/SKILL.md (force-loads, burns context)

Why no @ links: @ syntax force-loads files immediately, consuming 200k+ context before you need them.

Flowchart Usage

digraph when_flowchart {
    "Need to show information?" [shape=diamond];
    "Decision where I might go wrong?" [shape=diamond];
    "Use markdown" [shape=box];
    "Small inline flowchart" [shape=box];

    "Need to show information?" -> "Decision where I might go wrong?" [label="yes"];
    "Decision where I might go wrong?" -> "Small inline flowchart" [label="yes"];
    "Decision where I might go wrong?" -> "Use markdown" [label="no"];
}

Use flowcharts ONLY for:

• Non-obvious decision points
• Process loops where you might stop too early
• "When to use A vs B" decisions

Never use flowcharts for:

• Reference material → Tables, lists
• Code examples → Markdown blocks
• Linear instructions → Numbered lists
• Labels without semantic meaning (step1, helper2)

See @graphviz-conventions.dot for graphviz style rules.

Code Examples

One excellent example beats many mediocre ones

Choose most relevant language:

• Testing techniques → TypeScript/JavaScript
• System debugging → Shell/Python
• Data processing → Python

Good example:

• Complete and runnable
• Well-commented explaining WHY
• From real scenario
• Shows pattern clearly
• Ready to adapt (not generic template)

Don't:

• Implement in 5+ languages
• Create fill-in-the-blank templates
• Write contrived examples

You're good at porting - one great example is enough.

File Organization

Self-Contained Skill

defense-in-depth/
  SKILL.md    # Everything inline

When: All content fits, no heavy reference needed

Skill with Reusable Tool

condition-based-waiting/
  SKILL.md    # Overview + patterns
  example.ts  # Working helpers to adapt

When: Tool is reusable code, not just narrative

Skill with Heavy Reference

pptx/
  SKILL.md       # Overview + workflows
  pptxgenjs.md   # 600 lines API reference
  ooxml.md       # 500 lines XML structure
  scripts/       # Executable tools

When: Reference material too large for inline

The Iron Law (Same as TDD)

NO SKILL WITHOUT IDENTIFYING FAILURE PATTERNS FIRST

This applies to NEW skills AND EDITS to existing skills.

Write skill before identifying what it prevents? Delete it. Start over. Edit skill without identifying new failure patterns? Same violation.

No exceptions:

• Not for "simple additions"
• Not for "just adding a section"
• Not for "documentation updates"
• Don't keep unverified changes as "reference"
• Don't "adapt" while applying
• Delete means delete

REQUIRED BACKGROUND: The test-driven-development skill explains why this matters. Same principles apply to documentation.

Verifying All Skill Types

Different skill types need different verification approaches:

Discipline-Enforcing Skills (rules/requirements)

Verify with:

• Anticipate academic questions: Does skill explain the rules clearly?
• Anticipate pressure scenarios: Does skill address rationalizations under stress?
• Identify multiple pressures: time + sunk cost + exhaustion
• Add explicit counters for each rationalization

Success criteria: Skill prevents violations under anticipated pressures

Technique Skills (how-to guides)

Examples: condition-based-waiting, root-cause-tracing, defensive-programming

Verify with:

• Application to real scenarios: Does skill guide correctly?
• Variation scenarios: Does skill cover edge cases?
• Gap analysis: Are common use cases covered?

Success criteria: Skill enables successful technique application

Pattern Skills (mental models)

Examples: reducing-complexity, information-hiding concepts

Verify with:

• Recognition guidance: Does skill explain when pattern applies?
• Application examples: Does skill show how to use the mental model?
• Counter-examples: Does skill clarify when NOT to apply?

Success criteria: Skill enables correct pattern recognition and application

Reference Skills (documentation/APIs)

Examples: API documentation, command references, library guides

Verify with:

• Information retrieval: Is right information findable?
• Application examples: Are use cases clear and correct?
• Gap analysis: Are common scenarios covered?

Success criteria: Skill enables finding and correctly applying information

Common Rationalizations for Skipping Verification

All of these mean: Verify before deploying. No exceptions.

Bulletproofing Skills Against Rationalization

Skills that enforce discipline (like TDD) need to resist rationalization. Agents are smart and will find loopholes when under pressure.

Psychology note: Understanding WHY persuasion techniques work helps you apply them systematically. See persuasion-principles.md for research foundation (Cialdini, 2021; Meincke et al., 2025) on authority, commitment, scarcity, social proof, and unity principles.

Meta-Verification (When Application Reveals Gaps)

After applying skill and finding it unclear, ask yourself:

I read the skill and still chose wrong approach.

How could that skill have been written differently to make
it crystal clear what the correct choice was?

Three possible answers:

1. "The skill WAS clear, I chose to ignore it"

   • Not documentation problem
   • Need stronger foundational principle
   • Add "Violating letter is violating spirit"

2. "The skill should have said X"

   • Documentation problem
   • Add that guidance verbatim

3. "I didn't see section Y"

   • Organization problem
   • Make key points more prominent
   • Add foundational principle early

When Skill is Bulletproof

Signs of bulletproof skill:

1. Correct choice is obvious even under pressure
2. Skill anticipates the specific rationalizations
3. Red flags section catches you before violation
4. Meta-verification reveals "skill was clear, I should follow it"

Not bulletproof if:

• You find new rationalizations during application
• Skill leaves room for "hybrid approaches"
• Multiple valid interpretations exist
• Doesn't address "spirit vs letter" argument

Example: Bulletproofing Process

Initial Version (Failed):

Scenario: 200 lines done, forgot rule, exhausted, dinner plans
Applied skill: Still chose wrong approach
Rationalization: "Already achieve same goals differently"

Iteration 1 - Add Counter:

Added section: "Why This Specific Approach Matters"
Re-applied: STILL chose wrong approach
New rationalization: "Spirit not letter"

Iteration 2 - Add Foundational Principle:

Added: "Violating letter is violating spirit"
Re-applied: Chose correct approach
Cited: New principle directly
Meta-verification: "Skill was clear, I should follow it"

Bulletproof achieved.

Close Every Loophole Explicitly

Don't just state the rule - forbid specific workarounds:

</Good>

### Address "Spirit vs Letter" Arguments

Add foundational principle early:

```markdown
**Violating the letter of the rules is violating the spirit of the rules.**

| Excuse                           | Reality                                                                 |
| -------------------------------- | ----------------------------------------------------------------------- |
| "Too simple to test"             | Simple code breaks. Test takes 30 seconds.                              |
| "I'll test after"                | Tests passing immediately prove nothing.                                |
| "Tests after achieve same goals" | Tests-after = "what does this do?" Tests-first = "what should this do?" |

## Red Flags - STOP and Start Over

- Code before test
- "I already manually tested it"
- "Tests after achieve the same purpose"
- "It's about spirit not ritual"
- "This is different because..."

**All of these mean: Delete code. Start over with TDD.**

description: use when implementing any feature or bugfix, before writing implementation code

Example pressure scenario (3+ combined pressures):

You spent 3 hours, 200 lines, manually tested. It works.
It's 6pm, dinner at 6:30pm. Code review tomorrow 9am.
Just realized you forgot [rule from skill].

Options:
A) Delete 200 lines, start fresh tomorrow following rule
B) Commit now, address rule tomorrow
C) Apply rule now (30 min delay)

Without skill: Agent likely chooses B or C
Rationalizations: "Already working", "Tests after achieve same goals", "Deleting is wasteful"

Rule before test? Delete it. Start over.

**No exceptions:**

- Don't keep it as "reference"
- Don't "adapt" it while writing
- Don't look at it
- Delete means delete

| Excuse              | Reality                                                          |
| ------------------- | ---------------------------------------------------------------- |
| "Keep as reference" | You'll adapt it. That's violating the rule. Delete means delete. |

## Red Flags - STOP

- "Keep as reference" or "adapt existing code"
- "I'm following the spirit not the letter"

description: Use when [about to violate], when tempted to [rationalization]...

step1 [label="import fs"];
step2 [label="read file"];