Writing Skills

Overview

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

Personal skills live in agent-specific directories (e.g., ~/.claude/skills for Claude Code, ~/.codex/skills for Codex, or custom agent directories)

You write test cases (pressure scenarios with subagents), watch them fail (baseline behavior), write the skill (documentation), watch tests pass (agents comply), and refactor (close loopholes).

Core principle: If you didn't watch an agent fail without the skill, you don't know if the skill teaches the right thing.

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.

What is a Skill?

A skill is a reference guide for proven techniques, patterns, or tools. Skills help future agent 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

TDD Concept	Skill Creation
Test case	Pressure scenario with subagent
Production code	Skill document (SKILL.md)
Test fails (RED)	Agent violates rule without skill (baseline)
Test passes (GREEN)	Agent complies with skill present
Refactor	Close loopholes while maintaining compliance
Write test first	Run baseline scenario BEFORE writing skill
Watch it fail	Document exact rationalizations agent uses
Minimal code	Write skill addressing those specific violations
Watch it pass	Verify agent now complies
Refactor cycle	Find new rationalizations → plug → re-verify

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 (required) + optional supporting files. Flat namespace.

Separate files for: Heavy reference (100+ lines), reusable tools. Keep inline: Principles, code patterns (<50 lines).

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 [triggers/symptoms] - [what it does, third person]
---
# Skill Name
## Overview (1-2 sentences), ## When to Use (symptoms, NOT to use)
## Core Pattern (before/after code), ## Quick Reference (table for scanning)
## Implementation (inline or link), ## Common Mistakes, ## Real-World Impact (optional)

Agent Search Optimization (ASO)

Critical for discovery: Future agents need to FIND your skill

1. Rich Description Field

Purpose: Agents read 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)

Quality	Example
BAD	For async testing (vague), I can help... (first person), setTimeout/sleep (tech-specific but skill isn't)
GOOD	Use when tests have race conditions... - replaces timeouts with condition polling (problem + solution)

2. Keyword Coverage

Use words agents 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
• ✅ testing-skills-with-subagents not subagent-skill-testing

4. Token Efficiency (Critical)

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

Target word counts by skill type:

• Bootstrap/Getting-started: <150 words each (loads in every session)
• Simple technique skills: <500 words (procedures, patterns, single concept)
• Discipline-enforcing skills: <2,000 words (TDD, verification, systematic debugging - need rationalization tables)
• Process/workflow skills: <4,000 words (multi-phase workflows with comprehensive templates)

Rationale: Complex skills need extensive rationalization prevention and complete templates. Don't artificially compress at the cost of effectiveness.

Techniques: Reference --help instead of documenting flags. Cross-reference other skills instead of repeating. Compress examples (42 words → 20 words). Don't repeat cross-referenced content.

Verify: wc -w skills/path/SKILL.md (check against word counts above)

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 test-driven-development
• ✅ Good: **REQUIRED BACKGROUND:** You MUST understand 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

Only for: Non-obvious decisions, process loops, "A vs B" choices. Never for: Reference (→tables), code (→blocks), linear steps (→lists). See graphviz-conventions.dot for conventions.

Code Examples

One excellent example in most relevant language. Complete, well-commented WHY, real scenario, ready to adapt. Don't: multi-language, fill-in-blank templates, contrived examples.

File Organization

Type	Structure	When
Self-Contained	skill/SKILL.md only	All content fits inline
With Tool	SKILL.md + example.ts	Reusable code, not narrative
Heavy Reference	SKILL.md + *.md refs + scripts/	Reference >100 lines

The Iron Law (Same as TDD)

NO SKILL WITHOUT A FAILING TEST FIRST

This applies to NEW skills AND EDITS to existing skills.

Write skill before testing? Delete it. Start over. Edit skill without testing? Same violation.

No exceptions:

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

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

Testing All Skill Types

Skill Type	Examples	Test With	Success Criteria
Discipline (rules)	TDD, verification	Pressure scenarios (time + sunk cost + exhaustion), academic questions	Agent follows rule under maximum pressure
Technique (how-to)	condition-based-waiting, root-cause-tracing	Application + variation + gap testing	Agent applies technique to new scenario
Pattern (mental model)	reducing-complexity	Recognition + application + counter-examples	Agent identifies when/how to apply
Reference (docs/APIs)	API docs, command refs	Retrieval + application + gap testing	Agent finds and applies info correctly

Common Rationalizations for Skipping Testing

Excuse	Reality
"Skill is obviously clear"	Clear to you ≠ clear to other agents. Test it.
"It's just a reference"	References can have gaps, unclear sections. Test retrieval.
"Testing is overkill"	Untested skills have issues. Always. 15 min testing saves hours.
"I'll test if problems emerge"	Problems = agents can't use skill. Test BEFORE deploying.
"Too tedious to test"	Testing is less tedious than debugging bad skill in production.
"I'm confident it's good"	Overconfidence guarantees issues. Test anyway.
"Academic review is enough"	Reading ≠ using. Test application scenarios.
"No time to test"	Deploying untested skill wastes more time fixing it later.

All of these mean: Test 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.

Close Every Loophole Explicitly

Don't just state rule - forbid specific workarounds:

• BAD: Write code before test? Delete it.
• GOOD: Add Delete it. Start over. + explicit No exceptions: list (don't keep as reference, don't adapt, don't look, delete means delete)

Address "Spirit vs Letter" Arguments

Add foundational principle early:

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

This cuts off entire class of "I'm following the spirit" rationalizations.

Build Rationalization Table

Capture rationalizations from baseline testing (see Testing section below). Every excuse agents make goes in the table:

| 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?" |

Create Red Flags List

Make it easy for agents to self-check when rationalizing:

## 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.**

Update CSO for Violation Symptoms

Add to description: symptoms of when you're ABOUT to violate the rule:

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

RED-GREEN-REFACTOR for Skills

Phase	Action
RED	Run pressure scenario WITHOUT skill → document choices/rationalizations verbatim
GREEN	Write skill addressing specific failures → verify agent complies
REFACTOR	Find new rationalizations → add counters → re-test until bulletproof

REQUIRED SUB-SKILL: Use testing-skills-with-subagents for pressure scenarios, pressure types, hole-plugging, meta-testing.

Anti-Patterns

Pattern	Example	Why Bad
Narrative	"In session 2025-10-03, we found..."	Too specific, not reusable
Multi-language	example-js.js, example-py.py	Mediocre quality, maintenance burden
Code in flowcharts	step1 [label="import fs"]	Can't copy-paste, hard to read
Generic labels	helper1, step3, pattern4	Labels need semantic meaning

STOP: Before Moving to Next Skill

After writing ANY skill, you MUST STOP and complete the deployment process.

Do NOT:

• Create multiple skills in batch without testing each
• Move to next skill before current one is verified
• Skip testing because "batching is more efficient"

The deployment checklist below is MANDATORY for EACH skill.

Deploying untested skills = deploying untested code. It's a violation of quality standards.

Skill Creation Checklist (TDD Adapted)

Use TodoWrite for each phase.

Phase	Requirements
RED	3+ pressure scenarios, run WITHOUT skill, document rationalizations verbatim
GREEN	Name (letters/numbers/hyphens), YAML frontmatter (<1024 chars), description starts "Use when...", third person, keywords, address baseline failures, one excellent example, verify compliance
REFACTOR	New rationalizations → add counters, build rationalization table, create red flags, re-test
Quality	Flowchart only if non-obvious, quick ref table, common mistakes, no narrative
Deploy	Commit and push, consider contributing PR

Discovery Workflow

How future agents find your skill:

1. Encounters problem ("tests are flaky")
2. Finds SKILL (description matches)
3. Scans overview (is this relevant?)
4. Reads patterns (quick reference table)
5. Loads example (only when implementing)

Optimize for this flow - put searchable terms early and often.

The Bottom Line

Creating skills IS TDD for process documentation.

Same Iron Law: No skill without failing test first. Same cycle: RED (baseline) → GREEN (write skill) → REFACTOR (close loopholes). Same benefits: Better quality, fewer surprises, bulletproof results.

If you follow TDD for code, follow it for skills. It's the same discipline applied to documentation.