computer-use

Use this skill when the user wants to automate desktop or browser tasks — clicking, typing, screenshotting, running GUI apps, or filling web forms.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

67%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-structured and actionable with concrete tool inputs and setup commands, but it lacks explicit sequenced workflow steps and validation checkpoints for what is a destructive, real-system-control skill, which caps workflow clarity. Conciseness and progressive disclosure are solid but not maximal due to the inlined full client example.

Suggestions

Add an explicit numbered workflow (e.g., 1. Screenshot to assess state, 2. Act, 3. Screenshot to verify result) with a validation/screenshot-after-action checkpoint to lift workflow_clarity above the destructive-skill cap of 3.

Trim the full Python 'Usage Pattern' client block to the essential tool definitions, or move it to a references file, to improve conciseness and progressive disclosure.

Replace [x, y] placeholders in the tool snippets with concrete coordinate examples so the code is fully copy-paste executable.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient with concrete tool/JSON snippets and a quick-reference table, but the 'Usage Pattern' full Python client block re-explains setup Claude likely already knows from the computer-use API, making it slightly more than lean.	4 / 5
Actionability	Provides concrete, mostly executable JSON tool inputs and a full Python client example, plus setup commands; the tool-input snippets are copy-paste ready, though coordinate placeholders like [x, y] keep it just short of fully executable for all cases.	4 / 5
Workflow Clarity	The implied sequence (screenshot first, then act) is present but there are no explicit numbered steps or validation checkpoints for risky real-system operations, and the rubric caps destructive/system-control skills without validation at 3.	3 / 5
Progressive Disclosure	Well-organized into clear sections (Tools Available, Usage Pattern, Setup, Safety, Quick Reference) with no bundle files to reference; minor gaps are the inlined full API example that could live in a reference file, keeping it just above the score-3 anchor.	4 / 5
	Total	15 / 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.

A strong, concise description with explicit trigger guidance and multiple concrete actions across the desktop and browser domains. Only minor overlap risk with adjacent form/browser skills keeps it from full marks on distinctiveness.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions — 'clicking, typing, screenshotting, running GUI apps, or filling web forms' — giving comprehensive coverage of the skill's capabilities, matching the score-5 anchor.	5 / 5
Completeness	Explicitly answers both 'what' (automate desktop/browser tasks via clicking, typing, screenshotting, GUI apps, web forms) and 'when' ('Use this skill when the user wants to...') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Natural user phrases like 'automate desktop or browser tasks', 'filling web forms', and 'running GUI apps' are exactly what a user would say, covering synonyms across both desktop and browser domains.	5 / 5
Distinctiveness Conflict Risk	The desktop/browser automation niche is fairly distinct with clear triggers, but 'filling web forms' and 'typing' have some overlap with general browser/form skills, so it falls just short of the minimal-conflict anchor.	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: ProwlrBot/prowlr-marketplace
Path: external/anthropic-official/computer-use/SKILL.md
Commit: 32f0050

Reviewed: 3 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.