What is writing-skills?

Ideal for AI agents that need writing skills is test-driven development applied to process documentation. writing-skills is an AI agent skill for writing skills is test-driven development applied to process documentation.

How do I install writing-skills?

Run the command: npx killer-skills add growingio/growingio-sdk-ios-autotracker/writing-skills. It works with Cursor, Windsurf, VS Code, Claude Code, and 19+ other IDEs.

What are the use cases for writing-skills?

Key use cases include: Applying Writing skills IS Test-Driven Development applied to process documentation, Applying NO SKILL WITHOUT A FAILING TEST FIRST, Applying This applies to NEW skills AND EDITS to existing skills. No exceptions.

Which IDEs are compatible with writing-skills?

This skill is compatible with Cursor, Windsurf, VS Code, Trae, Claude Code, OpenClaw, Aider, Codex, OpenCode, Goose, Cline, Roo Code, Kiro, Augment Code, Continue, GitHub Copilot, Sourcegraph Cody, and Amazon Q Developer. Use the Killer-Skills CLI for universal one-command installation.

Are there any limitations for writing-skills?

Frontmatter (only two fields). ONLY describe triggering conditions — NOT workflow summaries. "Skill is obviously clear, no need to test" Clear to you ≠ clear to agents.

writing-skills

GrowingIO Analytics SDK for iOS/macOS/tvOS/watchOS/visionOS. It covers analytics, cocoapods, data-analysis workflows. This AI agent skill supports Claude

SKILL.md

Readonly

Upstream Repository Material

The section below comes from the upstream repository. Use it as supporting material alongside the fit, use-case, and installation summary on this page.

Upstream Source

Writing Skills

Name: writing-skills
Availability: InStock
Author: growingio

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

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

The Iron Law

NO SKILL WITHOUT A FAILING TEST FIRST

This applies to NEW skills AND EDITS to existing skills. No exceptions.

What is a Skill?

Is: Reusable technique, pattern, or reference guide for future agent instances
Is NOT: Narrative about a one-time problem; project-specific convention (put in AGENTS.md)

Skill Types (two orthogonal dimensions)

Type	Description	Example
Technique	具体方法，有步骤可循	`test-driven-development`
Pattern	思维模型	`brainstorming`
Reference	查询式，结构化条目	—（本项目暂无 Reference 型 skill）

Label	Meaning	Requirement
Rigid	必须严格遵守	必须带 Rationalizations + Red Flags
Flexible	原则可按场景取舍	不需 Rationalizations

Directory Structure

.agents/skills/                       ← 唯一存储位置（编辑这里）
  <skill-name>/
    SKILL.md                          # 必需
    <supporting-file>.md              # 仅当 >100 行或为可复用 prompt 模板

.claude/skills → ../.agents/skills    ← 软链接，让 Claude Code 默认路径也能读到

只编辑 .agents/skills/ 下的源文件——软链接自动同步。不要在 .claude/skills/ 下新建真实目录（会覆盖软链接）。

SessionStart hook 直接读 .agents/skills/using-growingio-sdk-skills/SKILL.md（见 .claude/hooks/session-start.sh）。

Frontmatter (only two fields)

yaml
1---
2name: skill-name-with-hyphens
3description: Use when <triggering condition>
4---

Claude Search Optimization (CSO)

Description rules:

Start with Use when / Use before / Use after
ONLY describe triggering conditions — NOT workflow summaries
Keep under 500 characters

WHY: When a description summarizes workflow, Claude follows the description instead of reading the full skill body. Description with workflow = agent skips body.

SKILL.md Body Structure

markdown
1# <Skill Name>
2
3> **Type:** Technique | **Discipline:** Rigid
4
5## Overview
6
7## When to Use / Not to Use
8
9## Process / Checklist
10
11## Rationalizations (Rigid skills required)
12| Excuse | Reality |
13
14## Red Flags (Rigid skills required)
15
16## Related Skills

RED-GREEN-REFACTOR for Skills

RED: Baseline Test

Run pressure scenario WITHOUT skill. Document rationalizations verbatim.

GREEN: Write Minimal Skill

Address those specific rationalizations. Run same scenarios WITH skill.

REFACTOR: Close Loopholes

Agent found new rationalization? Add counter. Re-test.

Rationalizations

Excuse	Reality
"Skill is obviously clear, no need to test"	Clear to you ≠ clear to agents
"Description needs workflow context"	Workflow in description → agent skips body
"Just adding a section, no test needed"	Same Iron Law
"I'll test if problems emerge"	Test BEFORE, not after

Red Flags

"I'll write the skill first, test later" → Delete. Start with RED.
"This edit is too small to test" → No edit is too small.
"Description needs to explain what this skill does" → That's what the body is for.

Note on the symlink layout

Canonical source：.agents/skills/<skill-name>/SKILL.md
Claude Code 读取路径：.claude/skills/<skill-name>/SKILL.md（通过软链接解析）
SessionStart hook 路径：.agents/skills/using-growingio-sdk-skills/SKILL.md（见 .claude/hooks/session-start.sh）

这个布局让不同 CLI 工具（各自默认读 .agents/ 或 .claude/）共享同一份文件。不要在 .claude/skills/ 下创建真实目录——会覆盖软链接。

writing-skills — for Claude Code writing-skills, growingio-sdk-ios-autotracker, community, for Claude Code, ide skills, analytics, cocoapods, data-analysis, growingio, marketing-automation

About this Skill

Features

# Core Topics

Skill Overview

Core Value

Ideal Agent Persona

↓ Capabilities Granted for writing-skills

! Prerequisites & Limits

About The Source

Browser Sandbox Environment

⚡️ Ready to unleash?

FAQ & Installation Steps

? Frequently Asked Questions