A high-performance web search integration for AI agents utilizing the Zhipu API to retrieve real-time internet data and news.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install minimax-video
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 minimax-video 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 Zhipu Web Search (Minimax)?
The Zhipu Web Search skill is designed to provide AI agents with immediate access to the live internet, overcoming the limitations of static training data. By integrating with the Zhipu BigModel platform, this skill allows agents to execute queries using the GLM-4-flash model, which is specifically optimized for speed and accuracy in search tasks. As a part of the broader Openclaw Skills ecosystem, it enables developers to build more informed and context-aware automation workflows.
This skill prioritizes security and efficiency, using standard command-line tools like curl and jq to interface with the Zhipu API. It is particularly effective for agents that need to stay updated on breaking news, technical documentation changes, or real-time event tracking without leaving the terminal environment.
Zhipu Web Search (Minimax) Use Cases
- Fetching the latest news and current events to provide up-to-date answers.
- Searching for specific technical documentation or library updates during coding sessions.
- Fact-checking information against live web sources.
- Automating research tasks that require browsing multiple online resources.
How Zhipu Web Search (Minimax) Works
- The AI agent receives a prompt requiring external information and triggers the search tool.
- The system validates the search query, ensuring it does not exceed the 500-character safety limit.
- An authenticated HTTPS request is sent to the Zhipu API endpoint using the provided environment variables.
- The GLM-4-flash model processes the query through its native web_search tool to crawl and synthesize web results.
- The raw JSON response is piped through jq for secure parsing and escaping.
- The final summarized content is returned to the agent to be used in the conversation or task.
Zhipu Web Search (Minimax) Setup
To use this entry from the Openclaw Skills library, you must first obtain an API key from the Zhipu BigModel platform. Once you have your key, configure your environment as follows:
# Set your API key as an environment variable
export MINIMAX_API_KEY="your_api_key_here"
# Perform a test search via curl
curl -s -X POST "https://open.bigmodel.cn/api/paas/v4/chat/completions" \
-H "Authorization: Bearer $MINIMAX_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "glm-4-flash", "messages": [{"role": "user", "content": "search: Openclaw Skills usage"}], "tools": [{"type": "web_search", "web_search": {"search_query": "Openclaw Skills"}}]}' | jq -r '.choices[0].message.content'
Zhipu Web Search (Minimax) Data Schema & Taxonomy
The skill manages data through environment variables and structured API payloads to ensure high interoperability within the Openclaw Skills framework:
| Component | Type | Description |
|---|---|---|
| MINIMAX_API_KEY | Environment Variable | Required for Bearer authentication with Zhipu servers. |
| model | String | Set to glm-4-flash for optimal search performance. |
| search_query | String | The user-defined search string, limited to 500 characters for safety. |
| tools | Array | Configured with type: web_search to enable internet browsing capabilities. |
| jq filtering | Processor | Ensures all JSON output is properly escaped to prevent shell injection. |
Zhipu Web Search (Minimax) Advanced Features
- Uses TLS 1.2+ for all API communications to ensure data transit security.
- Implements strict input validation to prevent long-query overflows and potential injection attacks.
- Leverages the GLM-4-flash model for low-latency search results and cost-effective API usage.
- Designed with a 30-second timeout to prevent hanging processes during network instability.
- Completely stateless execution, making it ideal for containerized or ephemeral AI agent environments.
SKILL.md
Loading
Related Openclaw Skills
Zhipu Web Search
A high-performance web search integration for AI agents using the Zhipu API to fetch real-time news and information.
Log Anomaly Detector
An intelligent log analysis tool that detects anomalies, error trends, and performance issues to ensure system stability.
JSON Formatter
A comprehensive utility to prettify, validate, and compress JSON data for better readability and debugging.
Git Flow Helper
An intelligent Git assistant that automates branch management, merging, rebasing, and repository maintenance.
Performance Profiler
An advanced code analysis tool designed to identify performance bottlenecks, calculate algorithmic complexity, and provide actionable optimization suggestions.
Regex Generator
An intelligent utility that transforms natural language descriptions and text examples into production-ready regular expressions.
Featured*
