Gemini Skill
ReviewAudited by ClawScan on May 1, 2026.
Overview
This skill appears coherent and benign, but it uses your logged-in browser session to submit prompts to Gemini and return answers or images.
Before installing, confirm you are comfortable letting the skill use your logged-in Gemini browser session and send selected prompts to Gemini; do not submit sensitive information unless that is acceptable.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The skill may use the account, model access, and quotas available in the signed-in Gemini session.
The skill relies on a logged-in browser profile to operate Gemini. This is purpose-aligned, but actions occur under the user's Gemini/Google session.
使用 OpenClaw 内置浏览器，`profile="openclaw"`。 ... 校验登录态（头像/输入框可见）。
Use it only with the intended browser profile/account, and switch or sign out if you do not want Gemini actions tied to that session.
Any private or sensitive content in the prompt will be submitted to Gemini for processing.
The skill explicitly sends user-provided text or image prompts to the external Gemini website. This is expected for a Gemini integration, but it is still an external data flow.
将用户问题原样输入并发送。 ... 将用户提示词原样输入。
Avoid including secrets or sensitive personal/business information unless you are comfortable sharing it with Gemini.
The skill can automate Gemini UI interactions such as entering prompts, toggling image generation, and sending requests.
The helper exposes page automation functions to find UI elements, fill the prompt, and click controls. This is consistent with the skill's purpose and does not show hidden endpoints or unrelated actions.
window.GeminiOps = {probe, click, fillPrompt, selectors:S, version:'0.1.0'};Use the skill for intended Gemini tasks and review results if the Gemini UI changes or behaves unexpectedly.