pproenca avatar

Skill Authoring

If you're building skills for Claude Code or MCP tools, this gives you the actual engineering practices that make them work reliably. It's 46 rules across metadata design, description engineering, and trigger optimization, the stuff that determines whether your skill activates when users need it or just sits there. The core insight is solid: descriptions drive activation, so you need specific capabilities and keywords upfront, not buried in implementation details. Progressive disclosure is broken down into a three-level system (overview, references, execution) that keeps token usage reasonable. Most valuable for the trigger optimization and testing sections, they're specific about covering synonyms and negative cases. Think of it as the missing manual for why some skills work and others don't.

skills.sh installs
330
GitHub stars
222
last updated
Aug 21, 2026

Install

$ npx -y skills add pproenca/dot-skills --skill skill-authoring --agent claude-code

Installs into .claude/skills of the current project.

Files

AI Agent Skills Best Practices

Design and development guide for AI agent skills, including Claude Code skills and MCP tools. Contains 46 rules across 8 categories, prioritized by impact to guide skill creation, review, and optimization.

When to Apply

  • Creating new Claude Code skills or MCP tools
  • Writing or reviewing SKILL.md metadata and descriptions
  • Optimizing skill trigger reliability
  • Structuring content for progressive disclosure
  • Testing skill activation and behavior
  • Designing tool interfaces for agent workflows

Core Principles

1. Descriptions drive activation. Claude selects skills based on description matching against user intent. Include specific capabilities, trigger keywords, and negative cases. A skill with a vague description activates inconsistently or never.

2. Front-load critical instructions. Claude may truncate long content. Place non-negotiable rules in the first 100 lines. Bury important constraints at the end and they get ignored.

Sponsors

Details

Repository
github.com/pproenca/dot-skills
Categories
Testing & QA, Documentation, AI & Agent Building, Design & UI/UX
First seen
Jun 3, 2026

More from pproenca/dot-skills

All 100
1davila7 avatarpyhealthdavila7/claude-code-templatesComprehensive healthcare AI toolkit for developing, testing, and deploying machine learning models with clinical data. This skill should be used when working wi325skills.sh installs
2glebis avatarpresentation-generatorglebis/claude-skillsGenerate interactive HTML presentations with neobrutalism styling, ASCII art decorations, and Agency brand colors. Outputs HTML (interactive with navigation), P325skills.sh installs
3bmad-code-org avatarbmad-checkpoint-previewbmad-code-org/bmad-methodLLM-assisted human-in-the-loop review. Make sense of a change, focus attention where it matters, test. Use when the user says "checkpoint", "human review", or "324skills.sh installs
4sanky369 avatartweet-writersanky369/vibe-building-skillsWrite high-engagement tweets and X/Twitter threads, grounded in live research of what's currently working in the user's niche: researches viral examples via web322skills.sh installs
5customerio avatarciocustomerio/cliCustomer.io CLI - use for Customer.io, Journeys, CDP Pipelines, builder onboarding, sandbox testing, app integration, SDK setup, transactional messaging, billin319skills.sh installs
6patricio0312rev avatarsecure-headers-csp-builderpatricio0312rev/skillsImplements security headers and Content Security Policy with safe rollout strategy (report-only → enforce), testing, and compatibility checks. Use for "security318skills.sh installs