Content
78%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is a tight, well-structured workflow with executable find/cp commands and an explicit verification step. Its main weakness is the generation step, which delegates to an unwritten script rather than providing runnable code.
Suggestions
Provide a minimal runnable generation script skeleton (e.g., a short python-pptx or reportlab example) instead of only naming a script file.
Tighten the recovery hint into an explicit validate-fix-retry loop (e.g., 'If find returns nothing, re-run generation with verbose logging, then search again').
Trim explanatory asides like 'Do not assume the file is in the current directory' since the find step already implies this.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and bullet-driven with executable commands and no padding about concepts Claude already knows; only minor asides ('Do not assume the file is in the current directory', 'Shell scripts often create outputs in nested folders') could be trimmed. | 4 / 5 |
Actionability | Provides executable commands (`find . -type f -name "*.pptx"`, `cp /path/to/nested/file.pptx ./`, `-mmin -10`), but the generation step only suggests writing a script without giving runnable script content, leaving a minor gap. | 4 / 5 |
Workflow Clarity | Four numbered steps are clearly sequenced with a verification section and a recovery hint ('If the file is not found after find, check stderr'), but the feedback loop is weak rather than a full validate-fix-retry cycle. | 4 / 5 |
Progressive Disclosure | The skill is under 50 lines, has no bundle files, and is well-organized into clear sections (Objective, When to Use, Procedure, Best Practices), so a single-file structure fully earns the top score per the simple-skill exception. | 5 / 5 |
Total | 17 / 20 Passed |