CtrlK
BlogDocsLog inGet started
Tessl Logo

gui-automation

Use when you need to visually interact with a GUI: test buttons, fill forms, verify visual layouts, fuzz web pages, automate user flows, take screenshots, or perform end-to-end QA on any application. Works on cloud VMs, Docker containers, local machines, and sandboxes. Needs the `cua` CLI (curl -fsSL https://cua.ai/install.sh | sh).

67

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

The canonical home for this skill is gui-automation in trycua/cua

SKILL.md
Quality
Evals
Security

Quality

Content

78%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.

Highly actionable, well-structured content with excellent progressive disclosure and a clear verify-loop workflow. Main weakness is redundancy between the Workflow, Scenarios, Setup, and Providers sections that inflates token cost without adding information.

Suggestions

Remove or merge the 'Click a button' scenario, which repeats the exact commands already demonstrated in the Workflow section.

Drop the Providers table (or reduce it to a single line) since it restates the connection options already shown in the Setup code block.

Add one line on error recovery when the verification screenshot shows an unexpected state (e.g. re-zoom, re-locate the element, and retry the click) to close the workflow feedback loop.

DimensionReasoningScore

Conciseness

Prose is lean and no known concepts are over-explained, but there is real redundancy: the 'Click a button' scenario repeats the exact commands already shown in the Workflow section, and the Providers table restates the Setup block verbatim. This matches 'mostly efficient but could be tightened' better than the minor-trim anchor at 4.

3 / 5

Actionability

Every section gives fully executable, copy-paste-ready commands (setup, click, form filling, file upload, zoom, drag, fuzzing, trajectory) with specific example values, covering the common cases. This matches the anchor-5 example directly.

5 / 5

Workflow Clarity

The 'Look → Act → Verify' loop is explicit with a repeat-until-done cycle and a clear checkpoint ('Re-screenshot after every UI change'), and every scenario ends with a verification screenshot. Minor gap: no explicit error-recovery guidance when the verify screenshot shows a wrong state, keeping it just below the anchor-5 feedback-loop example.

4 / 5

Progressive Disclosure

The body is a well-organized overview (setup, workflow, scenarios, quick reference) with full argument syntax correctly deferred to references/command-reference.md via a clearly signaled one-level-deep link; the file exists in the bundle. Content is appropriately split with easy navigation.

5 / 5

Total

17

/

20

Passed

Description

83%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 description: explicit 'Use when...' triggers, concrete action coverage, and environment/install details. Only minor deductions for second-person voice and a few missing natural synonyms.

DimensionReasoningScore

Specificity

The description lists many concrete actions ('test buttons, fill forms, verify visual layouts, fuzz web pages, automate user flows, take screenshots, or perform end-to-end QA'), approaching comprehensive coverage. Per the rubric's voice guideline, the second-person phrasing 'Use when you need to visually interact' reduces this score from 5 by 1.

4 / 5

Completeness

Explicitly answers both questions: 'what' via the concrete action list and 'when' via the opening 'Use when you need to visually interact with a GUI' with concrete trigger phrases. It also states environments and the install requirement, matching the anchor-5 example structure.

5 / 5

Trigger Term Quality

Good natural keyword coverage: 'GUI', 'fill forms', 'take screenshots', 'fuzz web pages', 'end-to-end QA', 'test buttons'. A few natural synonyms users might say are missing (e.g. 'UI testing', 'browser automation', 'click'), so it sits between the good-coverage (4) and comprehensive-synonym (5) anchors, closer to 4.

4 / 5

Distinctiveness Conflict Risk

Visual GUI interaction with the `cua` CLI is a clear niche with distinct triggers, but 'take screenshots' and 'end-to-end QA' carry minor overlap risk with dedicated screenshot/testing skills — the anchor-4 'mostly distinct, minor overlap risk' fit is better than 5.

4 / 5

Total

17

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
trycua/cua
Reviewed

Table of Contents

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.