← 返回 Skills 市场
1681
总下载
4
收藏
5
当前安装
7
版本数
在 OpenClaw 中安装
/install taobao-image-search
功能描述
使用淘宝进行以图搜同款、候选比对和加购物车操作。用户提供商品图片并要求“搜同款/找类似款/比价/加入购物车”时使用。优先执行本地脚本（run-taobao-task.js）完成全流程；当脚本失败或页面结构变化时回退 browser 工具手动执行。
安全使用建议
This skill appears to do what it says: automate Taobao image search and add-to-cart using Playwright. Before installing/running: (1) Only run on a trusted machine or sandbox because the scripts will save your Taobao login cookies to verification-artifacts/taobao-storage-state.json and .pw-user-data-taobao/ — treat those files like passwords and never upload/share them. (2) If you want no persistence, delete those files after each run or avoid the auto-login path. (3) Review the included scripts yourself (they are present and human-readable) and ensure you are comfortable with local file writes. (4) Install Playwright from the official source (npm) as instructed. If you need a higher assurance review, provide the full (untruncated) verify-taobao-runner.js to inspect any remaining code paths for network calls or unexpected behavior.
功能分析
Type: OpenClaw Skill
Name: taobao-image-search
Version: 1.1.3
The skill bundle provides automated Taobao image search and 'add to cart' functionality using Playwright. It includes logic for automated login detection and session persistence, saving browser state locally to 'verification-artifacts/taobao-storage-state.json'. While handling session tokens is a sensitive operation, the code is transparent, lacks any evidence of data exfiltration or remote command execution, and includes explicit security warnings in SKILL.md and README.md regarding the privacy of the generated credential files.
能力评估
Purpose & Capability
Name/description match the included scripts: run-taobao-task.js orchestrates image upload, search, candidate selection, and add-to-cart; auto-login-taobao.js handles interactive login and saves browser storage state. There are no unrelated credentials, binaries, or cloud APIs requested.
Instruction Scope
SKILL.md instructs running the local Playwright scripts and falling back to a browser tool; runtime actions are limited to reading a provided local image, driving the Taobao site, and writing verification artifacts. The skill explicitly documents that it saves session cookies/storage to verification-artifacts/taobao-storage-state.json and .pw-user-data-taobao/. No instructions reference reading unrelated host files or sending data to external endpoints other than Taobao.
Install Mechanism
No opaque download/install spec in the registry; SKILL.md asks to install Playwright (npm and npx playwright install chromium). This is a standard, expected dependency for browser automation and matches the code use.
Credentials
The skill requests no environment variables or external credentials. It does persist browser session tokens and user-data locally (cookies/storageState) to implement automatic login — this is functionally necessary but creates sensitive artifacts that the SKILL.md correctly warns about.
Persistence & Privilege
always is false and the skill does not request permanent platform-wide privileges. It writes its own artifacts and user-data under the skill directory (verification-artifacts and .pw-user-data-taobao) but does not modify other skills or system-wide configs.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install taobao-image-search - 安装完成后，直接呼叫该 Skill 的名称或使用
/taobao-image-search触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.1.3
No changes detected in this version (1.1.3).
- No file changes between this and the previous version.
- Functionality, usage, and documentation remain unchanged.
- Safe to update with no impact on skill behavior.
v1.1.2
Summary: This update adds privacy and environment safety notices, along with installation guidance.
- Added installation instructions for Playwright and Chromium in the skill metadata.
- Updated the "安全边界" section to "安全边界与隐私"，详细说明了会话数据本地存储、敏感性及清理方式，提醒用户妥善保管登录凭据。
- 建议仅在受信个人设备或沙箱环境下运行技能，增强了用户隐私保护与环境安全提示。
- 其余流程与逻辑未发生调整。
v1.1.1
- Removed the file save-taobao-cookie.js.
- No changes to functionality or documentation.
v1.1.0
update login detection
v1.0.2
No changes detected in this version.
- No file updates or modifications were made.
- Functionality, usage, and documentation remain unchanged.
v1.0.1
fix bugs
v1.0.0
Taobao Image Search Skill v1.0.0
淘宝图像搜索
- Enables searching for similar items, comparing, and adding items to cart on Taobao 淘宝 using images.
- Prioritizes local script automation; falls back to browser tool if scripts fail or site changes.
- Ensures user login is valid before proceeding.
- Only searches or adds to cart; ordering and payment are not performed.
- Asks users for missing key details (e.g., budget, purchase preference) if required.
元数据
常见问题
Taobao Image Search 是什么？
使用淘宝进行以图搜同款、候选比对和加购物车操作。用户提供商品图片并要求“搜同款/找类似款/比价/加入购物车”时使用。优先执行本地脚本（run-taobao-task.js）完成全流程；当脚本失败或页面结构变化时回退 browser 工具手动执行。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 1681 次。
如何安装 Taobao Image Search？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install taobao-image-search」即可一键安装，无需额外配置。
Taobao Image Search 是免费的吗？
是的，Taobao Image Search 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Taobao Image Search 支持哪些平台？
Taobao Image Search 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Taobao Image Search？
由 XR Gunner（@lazygunner）开发并维护，当前版本 v1.1.3。
推荐 Skills