Content
75%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.
A well-built, information-dense skill body: no filler, a compact tool reference, and one clear feedback-looped workflow. It falls short of top marks mainly on actionability (no per-tool argument examples) and workflow clarity (no error-recovery guidance), with mild conciseness and organization trim opportunities.
Suggestions
Add one-line usage examples showing argument shapes for the key tools, e.g. stave_lens_click("button#submit"), stave_lens_get_html("nav.sidebar"), and stave_lens_evaluate("document.title").
Add brief error-recovery guidance to the workflows, such as what to do when navigation fails, a selector matches nothing, or the console buffer shows errors after a click.
Trim agent-irrelevant detail (the WebContentsView backing, the Globe icon / Command Palette bullet) and consider moving the full session-presentation rules into a reference file, keeping only the decision rule inline.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and operational with no generic-concept padding, but a few tokens could be trimmed, such as the Electron "WebContentsView" implementation detail and the user-facing "Globe icon or Command Palette" bullet that adds little for the agent. | 4 / 5 |
Actionability | Concrete MCP tool names, required parameters (workspaceId), and an executable workflow example (stave_lens_navigate("http://localhost:3000/page")) provide mostly executable guidance, but per-tool invocation examples with argument shapes (e.g., selector syntax for stave_lens_click or stave_lens_inspect) are missing. | 4 / 5 |
Workflow Clarity | The Figma comparison workflow is a clear 6-step sequence with a validation step (compare rendered output to spec) and a feedback loop ("Repeat from step 2"), and the element picker is a numbered sequence, but there are no explicit error-recovery checkpoints (e.g., what to do when a tool call or navigation fails). | 4 / 5 |
Progressive Disclosure | Well-organized single-file skill with clear section headers and no nested references, but at ~95 lines the 15-row tool catalog and detailed session-presentation rules could partially live in a reference file; structure is good with minor organization gaps. | 4 / 5 |
Total | 16 / 20 Passed |