Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Deprecated — please use nyx-archive-coding-philosophy instead
What This Skill Does
A two-mode coding philosophy distinguishing feeling-first (discovery, expression) from structure-first (refactoring, maintenance). Includes a practical checklist derived from reviewing ~1800 lines of game code built across 10 iterations. Rooted in three principles: fallibilism, relational ontology, and play.
Gives a concrete decision framework for the tension between exploratory and production code rather than prescribing one style for all contexts.
When to use it
- Deciding when a prototype is ready to refactor
- Auditing dead code after a long creative session
- Extracting helper functions from repeated patterns
- Optimizing hot-path performance without premature abstraction
- Organizing a messy codebase by concern after building by intuition
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: I just finished a prototype with ~1800 lines. It works but the code is a mess. How do I clean it up?
- 1Confirm the user is ready to shift from feeling-mode to structure-mode
- 2Run an audit pass: read everything before changing, note dead code, duplicates, and hot paths
- 3Move all data/config to the top, all logic below it
- 4Extract any pattern appearing 3+ times into a named helper function
- 5Cache DOM lookups and object searches that run in loops
A prioritized refactoring plan organized by concern, with specific hotpath and dead-code findings called out