Content
32%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill reads more like a general reference page or README than an actionable skill for Claude. It spends significant tokens on capability listings and generic best practices that Claude already knows, while lacking validation steps, error handling, and clear sequential workflows. The actionable parts (CLI conversion commands) are useful but buried among verbose, low-value content.
Suggestions
Remove the 'When to Use This Skill', 'Core Capabilities', and 'Best Practices' sections entirely—they describe things Claude already knows and consume tokens without adding actionable value.
Add explicit validation steps to workflows: after conversion, verify the output file exists and has non-zero size; after UNO drawing creation, verify the file was saved successfully.
Restructure the UNO workflow to include the prerequisite step of starting LibreOffice in listening mode (`soffice --headless --accept=...`) as step 1, before attempting to connect.
Split detailed UNO API examples and troubleshooting into separate bundle files, keeping SKILL.md as a concise overview with clear references to those files.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is noticeably verbose with sections like 'When to Use This Skill' and 'Core Capabilities' that enumerate obvious capabilities Claude already knows. The 'Best Practices' section contains generic advice ('Use vector formats for scalability') that adds no value. The bulleted capability lists are padded and could be eliminated entirely. | 2 / 5 |
Actionability | The conversion commands are concrete and executable, and the UNO Python example is real code. However, the UNO example requires a running LibreOffice instance on a socket (which is only mentioned in troubleshooting, not in the workflow), the PNG export filter syntax may not work as shown, and many sections (Graphics Manipulation, Diagram Automation) are purely descriptive bullet lists with no executable guidance. | 3 / 5 |
Workflow Clarity | The workflows lack sequencing and validation. The UNO workflow doesn't mention the prerequisite of starting LibreOffice in listening mode. The batch conversion loop has no error handling or validation. There are no checkpoints to verify conversion success, no feedback loops for failures, and the 'Creating a New Drawing' workflow is incomplete—it doesn't show how to actually add content to the drawing. | 2 / 5 |
Progressive Disclosure | No bundle files exist, yet the skill is over 150 lines with content that should be split out (UNO API details, format conversion reference, troubleshooting). Everything is inlined in a single monolithic file with no references to supporting documents. The external resource links at the bottom point to general documentation rather than curated skill-specific references. | 2 / 5 |
Total | 9 / 20 Passed |