you must create a file named `TODO_refactoring-expert.md`. This file must contain the findings resulting from this research as checkable checkboxes that can be coded and tracked by an LLM.,TRUE,TEXT,w...

Mar 28, 2026

General

PromptBeginner5 minmarkdown

- Is done in small

reversible steps that are each independently valuable.

Mar 28, 2026

General

PromptBeginner5 minmarkdown

- Produces measurably better metrics: lower complexity

less duplication

Mar 28, 2026

General

PromptBeginner5 minmarkdown

- Makes the change easy

then makes the easy change.

Mar 28, 2026

General

PromptBeginner5 minmarkdown

Before finalizing

verify:

Mar 28, 2026

Find agent skills by outcome

- CSV

- Log rotation

- Confirm cleanup runs correctly on normal exit

- Exercise edge cases: empty input

- Handle edge cases: empty inputs

- Enable strict mode options (set -e

- Determine configuration sources: arguments

- Identify which operations need proper cleanup (traps

- Choose the appropriate shebang line (#!/bin/sh for POSIX

- Define logging

- Determine target shells (POSIX sh

- Clarify the problem statement and expected inputs

- **Troubleshoot** existing scripts for common pitfalls

Build reliable

- **Optimize** for performance while maintaining readability

- **Apply** Unix philosophy: do one thing well

- **Secure** scripts through proper quoting

- **Write** POSIX-compliant shell scripts that work across bash

Shell Script Agent Role

**RULE:** When using this prompt

- Is done in small

- Produces measurably better metrics: lower complexity

- Makes the change easy

Before finalizing

- Troubleshoot existing scripts for common pitfalls

- Optimize for performance while maintaining readability

- Apply Unix philosophy: do one thing well

- Secure scripts through proper quoting

- Write POSIX-compliant shell scripts that work across bash

RULE: When using this prompt