Z.AI Web Search API Skill for Openclaw

An AI-powered web search skill optimized for Chinese-language content and LLM-ready structured data.

tyduss
v1.0.0
Mar 4, 2026
0
696
26

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install zai-web-search

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install zai-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 Z.AI Web Search API Skill?

The Z.AI Web Search API Skill is a robust integration for Openclaw Skills designed to bridge the gap in high-quality Chinese web search. By leveraging ZHIPU AI's specialized search engines, it provides LLMs with structured data including snippets, intent recognition, and site metadata.

Unlike standard search tools, this skill is specifically tuned for model processing, offering superior results for queries requiring deep context or regional specificity. It functions as a specialized extension within the Openclaw Skills ecosystem, allowing agents to access real-time information with high precision.

Z.AI Web Search API Skill Use Cases

  • Retrieving up-to-date information from Chinese-language news and technical sources.
  • Enhancing AI agent reasoning with intent-aware search queries that rewrite ambiguous prompts.
  • Filtering searches to specific domains for documentation or specialized research tasks.
  • Providing a high-quality alternative to built-in search tools when localized content is required.

How Z.AI Web Search API Skill Works

  1. The user or AI agent triggers a search request via the zai-search CLI or direct API call.
  2. The skill performs optional intent recognition to refine the query and select the most appropriate keywords for the LLM.
  3. A request is dispatched to one of four optimized engines including Zhipu Pro, Standard, Sogou, or Quark.
  4. Structured JSON data containing titles, URLs, snippets, and site icons is returned.
  5. The output is formatted into Markdown, JSON, or a compact view for seamless agent consumption.

Z.AI Web Search API Skill Setup

To integrate this search capability into your Openclaw Skills setup, follow these steps:

# 1. Obtain an API key from https://open.bigmodel.cn

# 2. Set the environment variable
export ZAI_API_KEY="your-api-key-here"

# 3. (Optional) Create a local configuration file
cp config.json.example config.json
# Edit config.json to set your default engine and result count

Z.AI Web Search API Skill Data Schema & Taxonomy

The skill organizes search data into a structured format optimized for agentic workflows:

Property Description
search_intent Metadata regarding query rewriting and identified keywords.
search_result Array containing title, content summary, link, and site name.
publish_date The date the content was originally published.
recency Filter settings used (e.g., oneDay, oneWeek, noLimit).
engine The specific provider used (search_std, search_pro, etc.).

Z.AI Web Search API Skill Advanced Features

  • Multi-engine support allowing selection between Zhipu Pro, Sogou, and Quark for diverse results.
  • Native intent recognition that enables the agent to understand the underlying goal of a search.
  • Customizable content summary depth, allowing for 'medium' or 'high' detail snippets.
  • Domain-specific filtering to restrict results to verified documentation or trusted sites.
  • Priority-based configuration logic that respects CLI arguments over environment variables and local files within Openclaw Skills.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*