// SCAN REPORT

Gemini Skill

21 tools. 11 can modify or destroy data without limits.

11 write tools that can modify data. Rate limits recommended.

Last updated: 10 Jun 2026

11 can modify or destroy data

10 read-only

21 tools total

Community server · catalogue entry verified 10/06/2026

How to control Gemini Skill ↓

TOOL MAP

Read (10) Write / Execute (11) Destructive / Financial (0)

MOST DANGEROUS TOOLS

High Risk

Execute · Medium gemini_navigate_to Navigation via CDP is a browser action that executes external operations (changing page state, loading content, potentially triggering side effects like authentication flows or API calls). Execute · Medium gemini_new_chat This tool executes a browser control action via CDP that creates a new session/conversation state in the Gemini service. Execute · Medium atlas_send_message This tool triggers an external API call to Atlas Cloud, sending a message and receiving a response. Execute · High gemini_generate_image This tool triggers an external operation — it drives a real browser session via CDP to interact with the Gemini web interface and generate images.

11 of Gemini Skill's 21 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

HOW TO CONTROL GEMINI SKILL

PolicyLayer is an MCP gateway — it sits between your AI agents and Gemini Skill, and nothing reaches the server without passing your rules. These are the rules we recommend:

Rate limit write operations

{
  "gemini_upload_images": {
    "limits": [
      {
        "counter": "gemini_upload_images_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "atlas_list_models": {
    "limits": [
      {
        "counter": "atlas_list_models_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Gemini Skill — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON GEMINI SKILL →

Free to start. No card required.

ALL 21 GEMINI SKILL TOOLS

EXECUTE 10 tools

Execute gemini_navigate_to 打开指定的 Gemini 页面 URL（如特定会话链接）。仅允许 gemini.google.com 域名，其他域名会被拒绝。适用于需要恢复到某个历史会话继续对话的场景 Execute gemini_new_chat 在 Gemini 中新建一个空白对话 Execute atlas_stream_message 通过 Atlas Cloud 流式接口验证 Streaming 能力，并在服务端聚合后返回完整文本 Execute gemini_reload_page 刷新 Gemini 页面（页面卡住或状态异常时使用） Execute gemini_temp_chat 进入 Gemini 临时对话模式（不保留历史记录，适合隐私场景）。注意：临时会话按钮仅在空白新会话页面可见，本工具会自动先新建会话再进入临时模式 Execute atlas_send_message 通过 Atlas Cloud OpenAI 兼容接口发送一轮文本对话，返回最终文本结果 Execute gemini_generate_image 调用后台的 Gemini 浏览器会话生成高质量图片。【重要：长耗时工具】 - 本工具为同步阻塞调用，内部会等待 Gemini 生成完毕后才返回最终结果（成功/失败+文件路径）。 - 典型耗时 60~120 秒，复杂图 Execute gemini_send_message 向 Gemini 发送文本消息并等待回答完成（不提取图片，纯文本交互）。【长耗时工具】同步阻塞等待 Gemini 回复完毕才返回。典型耗时 10~60 秒，必须等到最终结果再回传用户。【返回值】直接返回 Gemin Execute gemini_set_thinking_depth 切换 Gemini 思考深度（standard 标准 / extended 扩展）。仅在支持的模型下生效 Execute gemini_switch_model 切换 Gemini 模型（pro / flash / flash-lite）

WRITE 1 tools

Write gemini_upload_images 向 Gemini 当前输入框上传图片（仅上传，不发送消息），可配合 gemini_send_message 组合使用

READ 10 tools

Read atlas_list_models 获取 Atlas Cloud 当前 API Key 可见的模型列表（OpenAI 兼容接口） Read gemini_browser_info 获取 Gemini 浏览器会话的连接信息（CDP 端口、WebSocket 地址、Daemon 状态等），方便外部工具直连浏览器 Read gemini_check_login 检查当前 Gemini 页面是否已登录 Google 账号 Read gemini_download_full_size_image 下载完整尺寸的图片（高清大图）。默认下载最新一张，也可通过 index 指定第几张（从0开始，从旧到新排列）。【长耗时工具】需要 hover 触发工具栏 + CDP 拦截下载，典型耗时 10~30 秒。必须等到最终结 Read gemini_extract_image 提取指定图片的 base64 数据并保存到本地文件。可从 gemini_get_images 获取图片 src URL Read gemini_get_all_text_responses 获取当前 Gemini 会话中所有文字回复内容（仅文字，不含图片等其他类型回复） Read gemini_get_images 获取当前 Gemini 会话中所有已加载的图片列表（不下载，仅返回元信息） Read gemini_get_latest_text_response 获取当前 Gemini 会话中最新一条文字回复（仅文字，不含图片等其他类型回复） Read gemini_probe 探测 Gemini 页面各元素状态（输入框、按钮、当前模型等），用于调试和排查问题 Read gemini_share_latest_image 为当前会话中的图片创建 Gemini 公共分享链接并直接返回该链接。默认分享最新一张图片，也可通过 index 指定历史图片（从0开始，从旧到新排列）。工具会自动点开 Share image，等待 Gemini 生

RELATED SERVERS

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

How do I prevent bulk modifications through Gemini Skill? +

The Gemini Skill server has 1 write tools including gemini_upload_images. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Gemini Skill.

How many tools does the Gemini Skill MCP server expose? +

21 tools across 3 categories: Execute, Read, Write. 10 are read-only. 11 can modify, create, or delete data.

How do I enforce a policy on Gemini Skill? +

Register the Gemini Skill MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Gemini Skill tool call.

Deterministic rules across all 21 Gemini Skill tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON GEMINI SKILL →

Free to start. No card required.

21 Gemini Skill tools catalogued and risk-classified — across an index of 42,500+ MCP servers.

Gemini Skill

// TOOL MAP

// MOST DANGEROUS TOOLS

// HOW TO CONTROL GEMINI SKILL

// ALL 21 GEMINI SKILL TOOLS

// RELATED SERVERS

// FAQ