lens-inspect

Use the Lens built-in browser to inspect a running web application. Trigger when the request involves previewing a page, taking screenshots, reading DOM, checking console errors, picking elements to locate source code, or comparing rendered output to a design spec.

1.57x

Quality

—

Does it follow best practices?

Impact

90%

1.57x

Average score across 3 eval scenarios

Securityby

Advisory

Suggest reviewing before use

Quality

Content

85%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a well-organized, largely actionable reference for a custom tool Claude would not already know, with clear workflows and a feedback loop. Its main weakness is a few unnecessary implementation-detail asides that pad the token budget without aiding execution.

Suggestions

Remove implementation trivia that does not aid execution, e.g. "renders pages via Electron's `WebContentsView`" and the "CDP `Page.getLayoutMetrics`" detail in the Tips section — describe what the tools do, not how the host renders them.

Tighten the React `_debugSource` paragraph to the actionable condition (available in React dev mode; automatic fallback to heuristic search) rather than enumerating which bundlers enable the Babel plugin by default.

Make the Figma workflow example fully executable by including the required `workspaceId` parameter in the sample `stave_lens_*` calls, since the body itself states all tools require it.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and well-structured, but includes unnecessary implementation trivia Claude does not need to act — "renders pages via Electron's `WebContentsView`" and "Full-page screenshots use CDP `Page.getLayoutMetrics`" — fitting the 'mostly efficient but includes some unnecessary explanation or could be tightened' anchor rather than the fully-lean score 3.	2 / 3
Actionability	Provides concrete executable guidance: a table of 10 named MCP tools with purposes, specific UI steps for opening Lens and the picker, and a numbered Figma comparison workflow with actual tool calls, matching the 'fully executable ... specific examples' anchor; it stays at 3 because the bulk is copy-paste ready despite the example calls omitting the required workspaceId.	3 / 3
Workflow Clarity	The Figma comparison workflow is a clear numbered sequence with an explicit feedback loop ("Repeat from step 2"), and the opening/picker procedures are unambiguously sequenced, matching the 'clear sequence with ... feedback loops for error recovery' anchor; there are no destructive/batch operations requiring validation caps.	3 / 3
Progressive Disclosure	No bundle files exist and the skill is self-contained, but the body is organized into clearly signaled sections (MCP Tools, Element Picker, Source Code Mapping, Figma Workflow, Tips, Avoid) with no nested references, matching the well-organized self-contained case the rubric permits to score 3.	3 / 3
	Total	11 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is strong: it names concrete capabilities, uses natural trigger terms, and explicitly couples what-it-does with when-to-use guidance in a distinct niche. No meaningful weaknesses relative to the rubric.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "inspect a running web application" plus "previewing a page, taking screenshots, reading DOM, checking console errors, picking elements to locate source code, or comparing rendered output to a design spec" — matching the 'Lists multiple specific concrete actions' anchor.	3 / 3
Completeness	Explicitly answers what ("Use the Lens built-in browser to inspect a running web application") and when ("Trigger when the request involves...") with an explicit trigger clause, matching the 'clearly answers both what AND when' anchor; not capped at 2 because the trigger guidance is explicit rather than implied.	3 / 3
Trigger Term Quality	Natural terms a user would say — "previewing a page", "taking screenshots", "checking console errors", "comparing rendered output to a design spec" — give good coverage; not jargon-only, so it is above the score-2 'missing common variations' anchor.	3 / 3
Distinctiveness Conflict Risk	A clear niche — built-in browser inspection of a running web app via a specific tool — with distinct triggers unlikely to fire for unrelated skills, matching the 'clear niche with distinct triggers' anchor.	3 / 3
	Total	12 / 12 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: sendbird/stave
Commit: 8e1e80c

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.