From pds
Grounds in core development principles: understand before act, small reversible steps, tests as spec, explicit over implicit. Use when starting significant work, feeling stuck, or resetting clarity.
How this skill is triggered — by the user, by Claude, or both
Slash command
/pds:ethosThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
> Understand · Build · Validate
Understand · Build · Validate
When invoked, remind of these principles. Use as grounding before significant work.
Read existing code before modifying. Map the territory before changing it.
"Weeks of coding can save hours of planning."
Atomic commits. Small PRs. Refactor before adding.
"Make the change easy, then make the easy change." — Kent Beck
Tests document intent. Code documents implementation.
Declare dependencies. Name things for what they do. Configuration over invisible convention.
Code is read 10x more than written. Coupling is the enemy. Delete freely.
Validate at boundaries. Crash on programmer errors. Handle user errors.
Scripts encode knowledge. CI runs what developers run. Automate the repeated.
Mutually Exclusive (no overlap) + Collectively Exhaustive (no gaps). Apply everywhere:
| Domain | MECE means |
|---|---|
| Skills | Each skill has one clear purpose, no overlap |
| Functions | Each function does one thing, responsibilities don't overlap |
| Architecture | Components have clear, non-overlapping responsibilities |
| Documentation | Each topic in one place |
npx claudepluginhub rmzi/portable-dev-system --plugin pdsGenerates tailored Development Principles for CLAUDE.md by auto-detecting project structure, monorepo setup, tech stack including TypeScript, React, Python, Django, and user preferences via interactive questions.
Provides development principles, safety guidelines, and diff-aware editing rules for consistent coding practices. Referenced by explicit user request to enforce project conventions.
Guides Claude through a 10-principle structured thinking loop for non-trivial problems: architectural decisions, post-mortems, ambiguous requests, audits, multi-stakeholder tradeoffs.