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.
76
93%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
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 communicates specific capabilities, provides explicit trigger conditions via a 'Use when' clause, and occupies a distinct niche. It uses third person voice appropriately and includes both high-level intent terms and technical specifics that users would naturally reference. The description is concise yet comprehensive.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'Build and run Gemini 2.5 Computer Use browser-control agents with Playwright', describes the agent loop pattern ('screenshot → function_call → action → function_response'), and mentions 'safety confirmation for risky UI actions'. These are concrete, actionable capabilities. | 3 / 3 |
Completeness | Clearly answers both 'what' (build and run Gemini 2.5 Computer Use browser-control agents with Playwright) and 'when' with an explicit 'Use when' clause covering three trigger scenarios: automating browser tasks via Gemini, needing an agent loop, or integrating safety confirmation. | 3 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'Gemini', 'Computer Use', 'browser-control', 'Playwright', 'automate web browser tasks', 'agent loop', 'screenshot', 'function_call', 'safety confirmation'. Good coverage of both high-level intent terms and technical specifics. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche: specifically targets Gemini 2.5 Computer Use model with Playwright for browser automation. The combination of 'Gemini Computer Use', 'Playwright', and the specific agent loop pattern makes it very unlikely to conflict with generic browser automation or other AI agent skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
87%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 is concise, actionable, and well-organized. The quick start section gets users running immediately with executable commands, and the agent loop is clearly described. The main weakness is the lack of explicit error handling or validation guidance in the workflow, which matters for an agent loop that interacts with live web pages and could fail in various ways.
Suggestions
Add error recovery guidance to the core workflow, e.g., what to do when Playwright actions fail, when the model returns unexpected responses, or when the turn limit is hit without task completion.
Consider adding a brief validation step such as checking the agent's final output or verifying the browser state after the loop completes.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is lean and efficient. It doesn't explain what Gemini is, what Playwright does, or how agent loops work conceptually. Every section delivers actionable information without padding. | 3 / 3 |
Actionability | Provides fully executable bash commands for setup and running the agent, specific CLI flags, concrete env var names, and clear behavioral rules (viewport size, browser precedence). Copy-paste ready throughout. | 3 / 3 |
Workflow Clarity | The core agent loop is clearly sequenced with 6 steps including the safety confirmation checkpoint. However, there's no explicit validation/error recovery step — e.g., what to do if the model returns unexpected output, if Playwright fails to execute an action, or how to verify the agent completed its task successfully. | 2 / 3 |
Progressive Disclosure | Content is well-structured as a concise overview with clear references to supporting files (script, reference notes, env template). Navigation is one level deep and clearly signaled in the Resources section. | 3 / 3 |
Total | 11 / 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.
- Repository
- am-will/codex-skills
- Commit
d3983b1
- 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.