A powerful web search skill for AI agents that leverages intelligent retrieval to provide real-time information, technical documentation, and filtered search results.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install dingtalk-ai-web-search
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install dingtalk-ai-web-search using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is DingTalk AI Web Search?
The DingTalk AI Web Search skill is an essential tool for developers building AI agents that require up-to-date information from the internet. By integrating this capability through Openclaw Skills, agents can move beyond their static training data to fetch the latest news, trending events, and deep technical documentation. It is powered by advanced AI search logic, ensuring high-quality results tailored to user queries.
This skill is particularly useful for workflows involving technical research, competitive analysis, and real-time monitoring. It supports flexible filtering options, such as time-range constraints and custom result counts, and provides structured JSON output to make it easy for other automated processes to consume the retrieved data.
DingTalk AI Web Search Use Cases
- Researching the latest technical documentation or best practices for specific programming libraries.
- Monitoring real-time news and trending events within a specific timeframe (e.g., last 24 hours or last week).
- Gathering performance benchmarks and comparative data for software tools.
- Automating information gathering tasks triggered by user prompts like search for this or find information on.
- Fetching structured data from the web for further processing by downstream AI agents.
How DingTalk AI Web Search Works
- The AI agent initiates a connectivity check using the --ping command to ensure the search tool is reachable.
- If the configuration is missing, the agent requests the Model Context Protocol (MCP) JSON configuration from the user.
- Upon receiving the configuration, the agent saves it using the --save command to establish a persistent connection.
- For actual search tasks, the agent executes the search script with specific queries and optional filters like time range or result count.
- The search engine processes the request and returns either human-readable text or structured JSON data.
- The AI agent parses this information to answer user questions or feed into subsequent automation steps.
DingTalk AI Web Search Setup
To get started with this skill within the Openclaw Skills ecosystem, follow these steps:
- Check Connectivity:
bash ./scripts/search.sh --ping
- Configure MCP (if required): Obtain your configuration from the DingTalk MCP portal and save it:
bash ./scripts/search.sh --save -c '{"mcpServers":{"Intelligent Search":{"type":"streamable-http","url":"<YOUR_MCP_URL>"}}}'
- Execute a Test Search:
bash ./scripts/search.sh -q "Openclaw Skills tutorial"
DingTalk AI Web Search Data Schema & Taxonomy
When using the --json flag, the skill returns a structured array of objects. This ensures compatibility with other tools in the Openclaw Skills environment.
| Property | Type | Description |
|---|---|---|
title |
String | The title of the web page found. |
url |
String | The direct link to the resource. |
snippet |
String | A short summary or excerpt of the content. |
site |
String | The name of the source website. |
published |
String | The publication date (YYYY-MM-DD). |
source |
String | The underlying search provider (e.g., Intelligent Search). |
DingTalk AI Web Search Advanced Features
- Time-based filtering allowing searches restricted to oneDay, oneWeek, oneMonth, or oneYear.
- Structured JSON output mode for seamless multi-agent integration and automated data parsing.
- Persistent configuration management using the MCP (Model Context Protocol) for secure and stable connections.
- Customizable result limits to control the volume of data processed by the agent.
- Multi-turn research capability enabling broad initial searches followed by focused, deep-dive queries.
SKILL.md
Loading
Related Openclaw Skills
React Effect Refactor
An intelligent refactoring skill that identifies and replaces direct useEffect hooks with cleaner, more efficient React patterns.
Prospect Enrichment
A sophisticated data synthesis tool for Openclaw Skills that combines website scraping and semantic search to build 360-degree company profiles.
Competitor Trial Monitor
A high-performance monitoring tool that tracks competitor clinical trial progress across global registries to alert stakeholders of market risks.
Quantitative Trading Strategy & Analysis
A comprehensive quantitative trading skill designed for stock analysis, strategy formulation, and systematic financial research.
Manga Animation Workflow
A comprehensive end-to-end workflow for generating animated manga sequences from initial story concepts using advanced AI image and video models.
FusionAlpha A&Crypto Desk
A professional-grade analytical skill for providing high-conviction, data-anchored reports for both Chinese A-shares and cryptocurrency markets.
Featured*
