Content
82%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.
An exemplary project-specific skill body: nearly every token is non-obvious repo knowledge, guidance is copy-paste executable, and workflows carry explicit guardrails. Minor room for improvement: drop the redundant '## Description' section, concretize the UI validation step in the add-a-string workflow, and consider offloading the schema-strings/hot-reload detail into reference files.
Suggestions
Remove the '## Description' section that restates the frontmatter description; the title plus frontmatter already establish scope, saving redundant tokens.
Make the 'Validate UI presentation' step in the add-a-string workflow concrete — e.g., name the exact check (run the affected preview, query the semantic tree via the compose-hot-reload MCP tools, or verify the string renders in the target pane).
Consider moving the schema-string generation rules (section 2, 'Schema strings are generated') and the desktop hot-reload MCP tooling (section 6) into a referenced file each, keeping SKILL.md as a leaner overview with one-level-deep pointers.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense with project-specific, non-obvious rules (deprecated API names, CMP format-string limitations, generated schema-string keying) and assumes Claude's competence throughout; the only trimmable instance is the opening '## Description' section, which restates the frontmatter — placing it at 'efficient; minor instances that could be trimmed' rather than the flawless 5. | 4 / 5 |
Actionability | Guidance is fully executable: a copy-paste-ready snippet with concrete outputs ('NumberFormatter.format(batteryLevel, 1) // "73.5"'), a decision table with real calls and results ('MetricFormatter.temperature(25.0f, isFahrenheit)' → '"77.0°F"'), exact commands ('python3 scripts/sort-strings.py', './gradlew :desktopApp:hotRun', './gradlew :schema-strings:sync') and exact file paths — matching the 'fully executable, covers common cases' anchor. | 5 / 5 |
Workflow Clarity | The add-a-string workflow is a numbered sequence with automated guardrails (':schema-strings:test' fails on bad edits; sort-strings.py regenerates the index), and the desktop MCP section has explicit checkpoints ('Poll status until connected: true before anything else', await_reload vs reload). It falls short of 5 only because step 4 'Validate UI presentation' does not specify how to validate. | 4 / 5 |
Progressive Disclosure | No bundle files exist and the single file is well-sectioned with clearly signaled references (strings-index.txt, 'see the navigation-and-di skill', an explicit Reference Anchors section with exact paths), giving good structure; the dense schema-strings rules and the hot-reload MCP tooling could arguably be split into separate reference files, leaving minor organization gaps versus the 5 anchor. | 4 / 5 |
Total | 17 / 20 Passed |