deepseek-web-query
使用 DeepSeek 网页版进行互联网查询，分担大模型请求和搜索负担。当用户需要查询最新信息、一般性知识、代码问题、文本分析等，或明确说"用 DeepSeek 查一下"、"联网搜索"、"查下最新"等时触发此技能。特别地，如果提问以"ds:"或"ds："开头，优先使用此技能。通过 Chrome DevTools MCP 控制浏览器与 DeepSeek 交互，自动检测登录状态并提示用户，保持浏览器会话复用，使用 evaluate_script 提取 Markdown 内容直接返回。
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/andeymei/deepseek-web-queryWhat This Skill Does
The deepseek-web-query skill serves as a high-performance bridge between the OpenClaw AI agent and the DeepSeek web interface. By utilizing the Chrome DevTools MCP (Model Context Protocol), this skill automates browser interactions to perform live internet searches, retrieve technical documentation, and access up-to-date information that may fall outside the agent's internal training data. It is engineered to maintain persistent browser sessions, intelligently detect login states, and extract markdown-formatted content directly from the DeepSeek chat environment, ensuring seamless information delivery without needing to manually copy-paste results.
Installation
To add this capability to your OpenClaw environment, execute the following command in your agent's terminal:
clawhub install openclaw/skills/skills/andeymei/deepseek-web-query
Ensure that you have the Chrome DevTools MCP and necessary browser dependencies configured, as the skill will automatically attempt to verify and repair these services upon initialization if they are detected as offline or misconfigured.
Use Cases
This skill is ideal for tasks requiring real-time web awareness. Use it when you need to research current market trends, verify the latest software versioning for complex development projects, or seek explanations for contemporary news stories. Because it handles the browser session for you, it is particularly effective for multi-step information gathering where you want to retain context throughout a dialogue with the DeepSeek model. It effectively delegates high-inference search tasks away from your main agent's compute budget.
Example Prompts
- "ds:最新的 Python 3.13 特性有哪些？"
- "用 DeepSeek 查一下今天关于人工智能的最新行业新闻"
- "搜索一下解决 React 内存泄漏的最新最佳实践并总结给我"
Tips & Limitations
- Pre-indexing: Always use the
ds:prefix for faster route triggering; the agent is specifically trained to prioritize requests starting with this marker. - Session Persistence: The skill leverages a dedicated profile (
openclaw); it is recommended to keep this browser session active to avoid redundant login procedures. - Login Handling: If the skill detects that you are logged out, it will pause and provide a screenshot. You must complete the login manually (e.g., via WeChat scan or OTP) within the browser window provided by the MCP session.
- Resource Management: Since this relies on real browser instances, ensure your host machine has sufficient RAM allocated for background Chrome processes, especially if running multiple concurrent agent tasks.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-andeymei-deepseek-web-query": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, code-execution