computer-use

Drive native desktop apps through DeepChat's built-in Computer Use tools. Use when the user asks to operate, inspect, automate, or perform a GUI task in a real desktop application.

Quality

91%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

96%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.

A lean, highly actionable, well-sequenced skill body with explicit validation and recovery loops befitting a destructive/automation context. Its only real weakness is that the linked reference files it points to do not exist in the bundle.

Suggestions

Ship the four referenced files (README.md, WEB_APPS.md, RECORDING.md, TESTS.md) or remove the dangling 'Linked References' entries so navigation does not break.

Confirm each in-body reference (e.g. 'Follow WEB_APPS.md', 'See ADVANCED-style notes') resolves to an actual bundled file before relying on it as a discovery path.

If a reference is intentionally external/runtime-provided rather than bundled, state that explicitly so progressive disclosure is not scored against a missing local file.

Dimension	Reasoning	Score
Conciseness	The body is dense and technical, assuming Claude's competence throughout — no basic explanations of what a GUI, app, or accessibility tree is; every section earns its place with actionable contract detail rather than padding.	5 / 5
Actionability	Provides fully executable, copy-paste-ready tool calls with exact parameters (e.g. start_session({ session, capture_scope: "auto" }), click({ pid, window_id, x, y, session })) covering the common cases across the action surface.	5 / 5
Workflow Clarity	The 9-step 'Required Loop' is a clear sequence with explicit validation checkpoints (steps 7-8 verify effect/postcondition) and feedback loops for error recovery (snapshot_id_required retry, escalate_session one-way transition), matching the checklist anchor.	5 / 5
Progressive Disclosure	Well-organized sections with a clearly signaled one-level 'Linked References' list, but the referenced files (README.md, WEB_APPS.md, RECORDING.md, TESTS.md) are not present in the bundle, so navigation would break — a structure gap that keeps it below 5.	4 / 5
	Total	19 / 20 Passed

Description

87%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, third-person description that answers both what and when with concrete, natural trigger terms and a clear niche. Only minor gaps in action specificity and synonym coverage keep it just below perfect on two dimensions.

Dimension	Reasoning	Score
Specificity	Names the domain ('native desktop apps') and several concrete actions ('operate, inspect, automate, or perform a GUI task'), with only minor coverage gaps; not quite the comprehensive multi-action list of a 5.	4 / 5
Completeness	Clearly states what it does ('Drive native desktop apps through DeepChat's built-in Computer Use tools') and an explicit when clause ('Use when the user asks to operate, inspect, automate, or perform a GUI task in a real desktop application') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes natural phrases users say ('operate', 'inspect', 'automate', 'GUI task', 'desktop application') with good coverage, but a few common variations are absent, short of the 5 anchor's synonym/extension comprehensiveness.	4 / 5
Distinctiveness Conflict Risk	The DeepChat Computer Use / native GUI automation niche is clearly scoped with distinct triggers, giving minimal overlap risk with other skills; uses third-person voice with no person penalties.	5 / 5
	Total	18 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: ThinkInAIXYZ/deepchat
Path: plugins/cua/skills/computer-use/SKILL.md
Commit: fa769d4

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.