Agent SkillsHashWarlock/nobody-plans-for-pi › nobody-writes-skills

nobody-writes-skills

GitHub

提供技能开发的TDD方法论,涵盖SKILL.md结构规范、前置条件描述规则及红绿重构测试流程,指导创建和验证Agent技能。

skills/nobody-writes-skills/SKILL.md HashWarlock/nobody-plans-for-pi

Trigger Scenarios

创建新的Agent技能 编辑现有技能文档 验证技能功能

Install

npx skills add HashWarlock/nobody-plans-for-pi --skill nobody-writes-skills -g -y
More Options

Use without installing

npx skills use HashWarlock/nobody-plans-for-pi@nobody-writes-skills

指定 Agent (Claude Code)

npx skills add HashWarlock/nobody-plans-for-pi --skill nobody-writes-skills -a claude-code -g -y

安装 repo 全部 skill

npx skills add HashWarlock/nobody-plans-for-pi --all -g -y

预览 repo 内 skill

npx skills add HashWarlock/nobody-plans-for-pi --list

SKILL.md

Frontmatter
{
    "name": "nobody-writes-skills",
    "description": "Use when creating new skills, editing existing skills, or verifying skills work before deployment"
}

Writing Skills

Overview

Writing skills IS Test-Driven Development applied to process documentation. Write test scenarios, watch agents fail without the skill, write the skill, watch them succeed.

Skill Structure

skill-name/
├── SKILL.md              # Required: frontmatter + instructions
├── scripts/              # Helper scripts (optional)
└── references/           # Detailed docs (optional)

SKILL.md Format

---
name: skill-name
description: Use when [specific triggering conditions]
---

# Skill Name

## Overview
Core principle in 1–2 sentences.

## When to Use
Symptoms and use cases.

## Process
Steps to follow.

## Common Mistakes
What goes wrong + fixes.

Frontmatter Rules

  • name: lowercase, hyphens, 1–64 chars, must match directory name
  • description: start with "Use when...", max 1024 chars, describe triggers NOT workflow

Critical: Description = WHEN to use, NOT what the skill does. Testing shows agents follow descriptions as shortcuts and skip the full skill body.

Testing Skills (TDD)

RED — Baseline

Run a pressure scenario with a subagent WITHOUT the skill. Document:

  • What choices did they make?
  • What rationalizations did they use?

GREEN — Write Skill

Write skill addressing the specific failures observed. Run same scenario WITH skill — agent should now comply.

REFACTOR — Close Loopholes

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

When to Create

Do create for: Reusable techniques, non-obvious patterns, cross-project workflows Don't create for: One-off solutions, project-specific config, mechanically enforceable rules

Checklist

  • Baseline scenario run (RED)
  • Skill addresses observed failures (GREEN)
  • Loopholes plugged (REFACTOR)
  • Name matches directory, description starts with "Use when..."
  • Tested with subagent

Version History

  • fc2edc0 Current 2026-07-25 05:06

Same Skill Collection

skills/nobody-brainstorms/SKILL.md
skills/nobody-debugs/SKILL.md
skills/nobody-executes-plans/SKILL.md
skills/nobody-finishes-a-development-branch/SKILL.md
skills/nobody-receives-code-review/SKILL.md
skills/nobody-requests-code-review/SKILL.md
skills/nobody-uses-git-worktrees/SKILL.md
skills/nobody-uses-tdd/SKILL.md
skills/nobody-verifies-before-completion/SKILL.md
skills/nobody-writes-plans/SKILL.md
skills/using-plans-for-nobody/SKILL.md

Metadata

Files
0
Version
fc2edc0
Hash
0c2f85c5
Indexed
2026-07-25 05:06

Главная - Вики-сайт
Copyright © 2011-2026 iteam. Current version is 2.155.2. UTC+08:00, 2026-08-20 08:22
浙ICP备14020137号-1 $Гость$