writing-skills
About
This skill provides structured templates and standards for creating or improving Claude agent skills. Use it when you need to build new skills (from simple to complex) or fix common issues like excessive length or poor AI adherence. It offers a decision tree to guide you to the right architecture and solutions.
Quick Install
Claude Code
Recommendednpx skills add boisenoise/skills-collections -a claude-code/plugin add https://github.com/boisenoise/skills-collectionsgit clone https://github.com/boisenoise/skills-collections.git ~/.claude/skills/writing-skillsCopy and paste this command in Claude Code to install this skill
GitHub Repository
Related Skills
compliance-evidence-pack
MetaThis skill helps developers assemble compliance evidence, logs, and audit artifacts. It's used when you need to generate required documentation for security, compliance, or governance reviews. The skill produces structured outputs like implementation plans, diagrams, specs, and validation checklists.
Agent Orchestrator
MetaThe Agent Orchestrator skill coordinates multiple specialized AI agents and skills to execute complex, multi-step workflows. It decomposes tasks, delegates to appropriate agents, manages dependencies, and synthesizes their outputs into a unified result. Use this meta-skill for projects requiring distributed expertise and coordinated execution across different capabilities.
architecture-review-and-approval
MetaThis Claude Skill runs structured architecture reviews to validate designs against requirements and constraints. It helps developers create implementation plans, required artifacts, and acceptance criteria with stakeholder sign-off. Use it when you need formal approval for security, compliance, or governance before proceeding with a build.
Agent Orchestrator
MetaThe Agent Orchestrator skill coordinates multiple specialized AI agents and tools to execute complex, multi-step workflows. It intelligently decomposes tasks, delegates to appropriate agents, and synthesizes their outputs into a coherent final result. Use this meta-skill for projects that require distributed expertise and managed inter-agent communication.