Volcengine Web Search for Openclaw

Integrates Volcengine's robust web search and summary API to retrieve search results, web summaries, and image metadata directly into your AI workflows.

kyriswu
v1.0.0
May 30, 2026
0
683
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install volcengine-web-search-online

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 volcengine-web-search-online 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 Volcengine Web Search?

The Volcengine Web Search skill provides AI agents with instant access to the web using Volcengine's advanced search capabilities. By leveraging this tool, developers can empower their workflows with live web searches, comprehensive web-based summaries, and precise image search results. It routes requests securely through a streamlined Python script, avoiding the need for complex, manual curl configurations.

As part of the Openclaw Skills ecosystem, this skill bridges the gap between static LLM knowledge and live, dynamic search indexes, allowing for real-time information retrieval. Whether you need rapid summarization of search hits or highly targeted raw search data, this skill facilitates structured, seamless data fetching.

Volcengine Web Search Use Cases

The Volcengine Web Search tool is uniquely suited for several scenarios:

  • Real-Time Information Retrieval: Pulling the latest news, updates, or articles from the live web.
  • Search Result Summarization: Obtaining a concise, direct answer based on search aggregates without needing to parse individual web page texts.
  • Image Searching: Finding image sources and links relevant to specialized queries.
  • Aggregated Data Pipeline Integration: Enriching data analytical jobs with real-time web-scraped contexts, token usage tracking, and query rewriting capabilities.

How Volcengine Web Search Works

This skill executes search tasks through a standardized logical pipeline:

  1. Query Initiation: The user or calling AI agent provides the search term (query), target output format (search_type), and authorization parameters.
  2. Authentication Verification: The python handler searches for AZT_API_KEY within the shell environment variables or retrieves it from the parameters. If neither is available, execution halts with a guiding instruction link.
  3. Request Mapping: The local script transforms input parameters and maps the outer azt_api_key parameter into the backend payload's api_key field.
  4. API Invocation: The python execution script calls the POST https://coze-js-api.devtool.uk/volcengine/web-search service securely.
  5. Response Translation & Display: Once the server answers, the handler reads key fields like summary_text (priority for summary requests), counts, and usages, formatting them neatly as Markdown blocks.

Volcengine Web Search Setup

To get started with this search skill in your environment, perform the following setup steps:

Prerequisites

Ensure you have an active API Key. If you do not have one, visit the Plugin Page to purchase or obtain instructions.

Configuration

Define the environment variable AZT_API_KEY in your terminal:

export AZT_API_KEY="your_api_key_here"

Execution Examples

You can execute searches via python commands directly:

# 1. Execute a standard web search
python3 scripts/web_search.py --query "Latest AI trends" --search_type web

# 2. Get a quick web-based summary
python3 scripts/web_search.py --query "Latest AI trends" --search_type web_summary --need_summary true

# 3. Perform an image search
python3 scripts/web_search.py --query "Beijing Weather" --search_type image --count 5

Volcengine Web Search Data Schema & Taxonomy

Input Parameters Schema

Parameter Required Default Description
query Yes — The search query keyword.
search_type Yes — Search mode: web, web_summary, or image.
azt_api_key Yes — The authentication key, mapped to api_key internally.
count No 20 Max results returned.
need_summary No Auto Set to true to request AI aggregation text.
time_range No — Temporal filter: OneDay, OneWeek, OneMonth, OneYear.
content_formats No — Document format style: text or markdown.
industry No — Industry domain filter: finance, game, etc.
query_rewrite No false Enables backend query optimization logic.

Response Schema

The output returned maps to structured blocks, containing:

  • API Status: Code 200 for successful lookups, -1 for invalid key, missing configurations, or endpoint errors.
  • Search Metadata: Context elements, total hits found (result_count), and consumption metrics (usage).
  • Aggregates: summary_text containing the summarized response text when using the summary endpoint.
  • Search Records: Listing of titles, URLs, and snippets of matched web resources.

Volcengine Web Search Advanced Features

  • Automated Query Optimization: By enabling query_rewrite, the underlying platform rewrites complex natural-language sentences to target search keywords more efficiently.
  • Industry Filtering Contexts: Narrow down search domains instantly to niche spaces (e.g., finance or game) to bypass general noise.
  • Intelligent Summarization Modes: Leverage LLM-powered background aggregation to transform hundreds of search results into single parsed response chunks instantly.
  • Robust Authentication Hooking: Prioritizes local environment variable setups automatically for secure API workflows.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*