CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

computer-use-macos

Operate macOS application UI through computer_use MCP and the Computer → App → Element SDK. Use when observing, clicking, or typing on a macOS MCP host, including a remote host. Select by target desktop, not the Agent machine.

Quality

85%

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

SKILL.md
Quality
Evals
Security

Quality

Content

85%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.

Excellent operational content: executable code, complete API signatures, explicit validation and error-recovery loops, and tight token efficiency. The only structural weakness is that everything lives in one file — the large API reference section would be better split into a references/ file with clear pointers.

Suggestions

Move the ~85-line typed API surface block into a references/api.md file and keep a short usage example inline in SKILL.md, linking to it explicitly.

Remove duplicated statements such as the app-activation note appearing in both the 'macOS environment' section and Workflow step 1.

Consider splitting the long 'Reading observations' bullet list into a reference file, keeping only the most common interpretation rules in the main body.

DimensionReasoningScore

Conciseness

The body is dense and operational with no basic-concept padding; nearly every sentence carries SDK-specific facts Claude cannot know. Not 5 because of minor redundancy (e.g., "The app may activate itself during launch" repeats the environment section) and a few over-wrapped caveats that could be tightened.

4 / 5

Actionability

Fully executable, copy-paste-ready examples ("app = computer.app(...); state = await app.read()", "await app.element(\"42\").typeText(\"hello\", {\"replace\": True})\") plus a complete typed API surface, concrete option dictionaries, error codes, and recovery guidance covering the common cases.

5 / 5

Workflow Clarity

Clear observe → act → verify sequence with explicit validation checkpoints ("read before making a decision that depends on changed UI", "Verify the intended document before editing", "Read explicitly afterwards to confirm the effect") and a full Results-and-recovery section with feedback loops driven by code/effect/next.

5 / 5

Progressive Disclosure

Section headers are clear, but the ~85-line inline API surface block is content that could live in a references/ file, matching 'some structure; content that should be separate is inline'. No bundle files exist, and the under-50-lines exception does not apply to a 190-line skill. Not 4 because there are no well-signaled external references at all; not 2 because the sections are well organized, not a wall of text.

3 / 5

Total

17

/

20

Passed

Description

78%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 with a clear what, an explicit 'Use when' clause, concrete action verbs, and highly distinctive platform/tool identifiers. Main gap is trigger breadth — a few natural terms (screenshot, desktop, remote control) would round it out.

DimensionReasoningScore

Specificity

Names the domain ("Operate macOS application UI") and several concrete actions ("observing, clicking, or typing") in third person, matching the 'lists several specific actions; minor gaps' anchor. Not 5 because coverage is incomplete (no scrolling, dragging, menu commands, or screenshots); not 3 because more than 1-2 actions are named.

4 / 5

Completeness

Explicitly answers both: what ("Operate macOS application UI through computer_use MCP and the Computer → App → Element SDK") and when ("Use when observing, clicking, or typing on a macOS MCP host, including a remote host"). Not 5 because the trigger set is narrow compared to the comprehensive-synonyms anchor; not 3 because the 'when' is explicit, not merely implied.

4 / 5

Trigger Term Quality

Includes natural terms users would say — "macOS", "clicking", "typing", "remote host" — matching 'good keyword coverage; a few natural terms missing'. Not 5 because common variations like "screenshot", "desktop", or "UI automation" are absent.

4 / 5

Distinctiveness Conflict Risk

Highly specific markers ("computer_use MCP", "Computer → App → Element SDK", "macOS MCP host") establish a clear niche with distinct triggers and minimal overlap risk with other skills.

5 / 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

XiaoLuoLYG/GOD

Path

jiuwenclaw/jiuwenswarm/cua_engine/skills/computer-use-macos/SKILL.md

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.