Guide authors in creating Claude Code skills with auto-activation and testing.
npx skills add https://github.com/fredpope/bench --skill skill-developer-fredpope
Please help me install this Agent Skill. Skill: skill-developer Source: https://github.com/fredpope/bench/tree/main/.claude/skills/skill-developer Command: npx skills add https://github.com/fredpope/bench --skill skill-developer-fredpope
This Skill helps knowledge workers and developers design, document, and maintain Claude Code skills with a clear, repeatable pattern for auto-activation, progressive disclosure, and robust testing.
The Quick Start section outlines how to draft a new skill: write SKILL.md, add an entry to skill-rules.json, and validate with manual hook tests.
High-intent search queries and answers about installing and using this skill.
To create a Claude Code skill with auto-activation, you draft a SKILL.md file with proper frontmatter, add a configuration entry to skill-rules.json, and validate the trigger patterns using manual hook tests to ensure predictable activation.
Trigger patterns in skill-rules.json include keywords, intent patterns, file paths, and content patterns. These trigger types work with configurable enforcement modes to control how and when a skill activates automatically.
The two-hook architecture ensures safe, automated skill activation by validating trigger conditions and enforcing rules before execution. This mechanism minimizes false positives and maintains predictable behavior during skill deployment.
Progressive disclosure in skill development is a documentation strategy that structures SKILL.md files to reveal information incrementally. It satisfies testing requirements by incorporating reference files and validation steps for robust skill maintenance.
You test skill frontmatter and enforcement levels by running manual hook tests to validate trigger patterns and auto-activation behavior. This process verifies that required fields are correctly structured and activation rules function as intended.
You should adjust enforcement modes for skill triggers when configuring auto-activation rules to balance strictness and flexibility. Proper enforcement configuration prevents false positives and ensures safe integration into the skill-rules.json architecture.
cogni-workInitialize a shared Cogni workspace with settings, plugin discovery, and metadata.
chacha95Create and manage Claude Code skills with SKILL.md and skill-rules.json.
shafibabarEnforce two-tier precedence for product-specific JSON config overrides against factory defaults.
infiquetraAutomates Python project setup with pyproject.toml, ruff, mypy, and pre-commit hooks.
