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.

1.02x

Quality

85%

Does it follow best practices?

Impact

92%

1.02x

Average score across 2 eval scenarios

Securityby

Low

Low-risk findings worth noting

Quality

Content

75%Weight 40%Scale 1-5

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

The body is well-organized, mostly concise, and provides concrete tool calls and a clear multi-step workflow with numbered steps. Weakest area is conciseness, where a few explanatory passages could be tightened.

Suggestions

Tighten conciseness: trim the 'What is Lens?' framing and the Electron WebContentsView backing detail, which explain internals Claude does not need to use the skill.

Improve actionability by listing the key parameters for each MCP tool (e.g., selector, fullPage/viewport flag for screenshot) so calls are copy-paste ready rather than just naming the tool.

Add an explicit validate/retry checkpoint to the Figma comparison workflow (e.g., 'If the screenshot still mismatches the spec after a fix, re-screenshot and re-compare before moving on').

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and assumes Claude's competence about browser/DOM concepts, but a few lines over-explain (e.g., the Electron WebContentsView backing detail and the 'Lens is a per-workspace embedded browser...' framing) that could be trimmed without losing actionable value.	4 / 5
Actionability	Concrete tool names, parameters ('All tools require a workspaceId'), a numbered Figma workflow with actual tool calls, and specific Avoid rules make the guidance mostly executable; minor gaps are that tool arguments beyond workspaceId and selector are not enumerated, so it is not fully copy-paste ready.	4 / 5
Workflow Clarity	Multi-step processes are clearly sequenced — the Figma design comparison workflow is a numbered loop, and the Element Picker is a 3-step numbered list — with most checkpoints present; it is not a 5 because there are no explicit validate/retry error-recovery steps, though the operations are non-destructive so the destructive-cap does not apply.	4 / 5
Progressive Disclosure	Good structure with clearly headed sections (Session Behavior, MCP Tools table, Element Picker, Source Code Mapping, workflow, Tips, Avoid) and no bundle files to split out; it is not a 5 only because the MCP Tools table and Tips inline content that could arguably live in a one-level-deep reference file, but as a self-contained ~95-line skill the organization is strong.	4 / 5
	Total	16 / 20 Passed

Description

96%Weight 40%Scale 1-5

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

This is a strong, well-structured description that clearly states both what the skill does and when to trigger it, with concrete natural-language trigger phrases. The only soft spot is minor overlap risk with generic browser inspection tooling.

Dimension	Reasoning	Score
Specificity	Names the domain (Lens built-in browser / inspecting a running web app) and lists multiple concrete actions: 'previewing a page, taking screenshots, reading DOM, checking console errors, picking elements to locate source code, or comparing rendered output to a design spec' — comprehensive coverage matching the score-5 anchor.	5 / 5
Completeness	Explicitly answers 'what' ('Use the Lens built-in browser to inspect a running web application') and 'when' via the 'Trigger when the request involves ...' clause with concrete trigger phrases, fully matching the score-5 anchor.	5 / 5
Trigger Term Quality	Uses natural user phrasing including 'previewing a page', 'taking screenshots', 'checking console errors', and 'comparing rendered output to a design spec' — these are phrases a user would naturally say, matching the comprehensive-synonym anchor at 5.	5 / 5
Distinctiveness Conflict Risk	The Lens-specific framing gives it a clear niche with distinct triggers and low conflict risk, but the generic inspection verbs (screenshots, console errors, DOM) could overlap with general browser/devtools skills; sits just below the clear-niche anchor of 5.	4 / 5
	Total	19 / 20 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
Path: skills/lens-inspect/SKILL.md
Commit: 69ca766

Reviewed: 5 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.