gemini-computer-use
Build and run Gemini 2.5 Computer Use browser-control agents with Playwright. Use when a user wants to automate web browser tasks via the Gemini Computer Use model, needs an agent loop (screenshot → function_call → action → function_response), or asks to integrate safety confirmation for risky UI actions.
Install with Tessl CLI
npx tessl i github:am-will/codex-skills --skill gemini-computer-use100
Does it follow best practices?
Validation for skill structure
Discovery
100%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 an excellent skill description that clearly specifies the technology stack (Gemini 2.5 Computer Use + Playwright), concrete capabilities (browser-control agents, agent loops, safety confirmation), and explicit trigger conditions. The description uses proper third-person voice and provides enough detail to distinguish it from other browser automation or AI agent skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'Build and run browser-control agents', 'automate web browser tasks', 'agent loop (screenshot → function_call → action → function_response)', and 'integrate safety confirmation for risky UI actions'. | 3 / 3 |
Completeness | Clearly answers both what ('Build and run Gemini 2.5 Computer Use browser-control agents with Playwright') and when ('Use when a user wants to automate web browser tasks via the Gemini Computer Use model, needs an agent loop, or asks to integrate safety confirmation'). | 3 / 3 |
Trigger Term Quality | Includes natural keywords users would say: 'Gemini 2.5', 'Computer Use', 'browser-control', 'Playwright', 'automate web browser tasks', 'agent loop', 'screenshot', 'safety confirmation', 'UI actions'. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with specific technology stack (Gemini 2.5, Computer Use, Playwright) and unique use case (browser automation via vision-based agent loop). Unlikely to conflict with generic browser or automation skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
100%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-crafted skill that efficiently teaches Claude how to build and run Gemini Computer Use agents. It excels at providing concrete, executable commands while maintaining brevity. The safety confirmation workflow is appropriately highlighted, and the structure allows for quick reference or deeper exploration.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is lean and efficient, with no unnecessary explanations of concepts Claude would already know. Every section serves a clear purpose without padding. | 3 / 3 |
Actionability | Provides fully executable bash commands for setup and running the agent, with specific flags and examples. The workflow steps are concrete and the browser selection options are clearly specified. | 3 / 3 |
Workflow Clarity | The core workflow section clearly sequences the agent loop with explicit steps including the safety confirmation checkpoint at step 4. The quick start provides a clear numbered sequence for setup. | 3 / 3 |
Progressive Disclosure | Well-organized with a quick start overview, then deeper topics in separate sections. References to external files (script, reference notes, env template) are clearly signaled and one level deep. | 3 / 3 |
Total | 12 / 12 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 — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- 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.