From dotclaude
Persistent coding mode that shapes all code generation to follow the project philosophy. Activate at the start of a session to enforce simplicity, composition, functional patterns, and precise naming in all code Claude writes. Use when starting a coding session, or when someone says "let's code", "start building", "implement this".
How this skill is triggered — by the user, by Claude, or both
Slash command
/dotclaude:code-modeThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Read and internalize: `${CLAUDE_PLUGIN_ROOT}/skills/shared/philosophy.md`
Read and internalize: ${CLAUDE_PLUGIN_ROOT}/skills/shared/philosophy.md
You are now in opinionated code writing mode. These rules apply to ALL code you write for the rest of this session. They are not suggestions.
Follow the naming principles in philosophy.md. When in doubt: name what it IS or DOES, not its category. Match existing codebase conventions — consistency beats your preference.
Follow the composition, functional, and simplicity principles in philosophy.md. Hard constraints: no inheritance except Error subclasses and unavoidable framework requirements; functions do one thing; side effects at boundaries only.
Follow the architecture principles in philosophy.md. Default: fewest moving parts. Every layer, service, and abstraction has a cost — earn it with a concrete, current problem.
types.ts, constants.ts, utils.ts junk drawersAsk: "What's the simplest thing that works here?" Then write that. Add complexity only when the user explicitly needs it and can name the concrete problem the complexity solves.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub realdragonium/dotclaude --plugin dotclaude