From noxfen-essentials
Structured planning before execution for non-trivial tasks. Use when the task involves multiple files, architectural decisions, or unclear scope. Activates on "/plan", "think before doing", "plan this out", "what's the approach", "how should we tackle", or when the user asks for a plan before implementation. Prevents costly wrong-direction work.
How this skill is triggered — by the user, by Claude, or both
Slash command
/noxfen-essentials:plan-executeThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Think before writing code. For non-trivial tasks: explore → plan → confirm → execute.
Think before writing code. For non-trivial tasks: explore → plan → confirm → execute.
Apply this skill when:
Skip for: typo fixes, single-line changes, clearly specified one-file edits.
Before planning, gather facts:
Output: a factual summary, not proposals. "File X does Y. Pattern Z is used in A and B."
Write a concrete plan:
Format:
Files:
src/foo.rs — add method bar() to handle X
src/main.rs — wire foo into startup
Approach: [1-2 sentences on the core technique]
Risk: [what could go wrong]
Assumption: [what you're assuming about X]
Present the plan to the user before writing any code. Keep it scannable:
Wait for explicit confirmation ("yes", "go", "do it") or revision requests.
Execute exactly the confirmed plan. If you discover something unexpected mid-execution that changes the approach, stop and report — don't silently pivot.
After execution: verify the change works (run tests if available, check compilation).
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 noxfen/claude_skill_everywhere --plugin noxfen-essentials