Content
46%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 content gives concrete, example-driven guidance but is weighed down by ALL-CAPS verbosity, redundant overlapping sections, a malformed primary example, and a mismatch between the informational-style title and the summarize-focused body. Tightening and de-duplicating would materially raise quality.
Suggestions
Resolve the title/body mismatch: the description promises an informational-style rewrite but the body repeatedly instructs summarization — pick one purpose and align the heading, chain-of-thought, and examples to it.
Fix the malformed EXAMPLE block (empty ``` fences at lines 64-65 and the stray ---) so the canonical before/after demonstration renders correctly and unambiguously.
Collapse the redundant CHAIN OF THOUGHT, TASK EXECUTION INSTRUCTIONS, and WHAT NOT TO DO sections into a single concise workflow, and drop the ALL-CAPS shouting and padded finale to improve token efficiency.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | ALL-CAPS restatement, redundant sections (CHAIN OF THOUGHT overlaps TASK EXECUTION INSTRUCTIONS and WHAT NOT TO DO), and a padded finale ("MAKE EVERY WORD COUNT. WRITE LIKE YOU SPEAK—") make it noticeably verbose with several padded passages. | 2 / 5 |
Actionability | Concrete do/don't guidance and bad/good word swaps ("Utilize"→"Use") plus few-shot examples provide real direction, but the primary EXAMPLE block is malformed with empty code fences and a stray "---", leaving a key detail broken. | 3 / 5 |
Workflow Clarity | A 7-step chain-of-thought sequence exists, but it competes with parallel TASK EXECUTION INSTRUCTIONS and WHAT NOT TO DO sections, making the actual execution sequence ambiguous despite being a simple non-destructive task. | 3 / 5 |
Progressive Disclosure | The body is self-contained with clear section headers and no external references, which is well-organized for a single-purpose skill, though it runs over 50 lines with some content that could be tightened. | 4 / 5 |
Total | 12 / 20 Passed |