← Back to Skills Marketplace
alexjunanjing-2

Search Web - CN

by AlexJu · GitHub ↗ · v0.1.0
cross-platform ✓ Security Clean
4150
Downloads
6
Stars
51
Active Installs
1
Versions
Install in OpenClaw
/install openclaw-skill-search-web
Description
Openclaw/Trae 联网搜索国内方案,使用火山引擎联网问答智能体 API 进行网络搜索问答。
Usage Guidance
Install only if you are comfortable sending search questions to Volcengine/feedcoopapi using your API credentials. Use a dedicated key if possible, monitor provider usage or billing, and do not include secrets, private source code, internal URLs, personal data, or regulated information in search prompts unless external sharing is approved.
Capability Analysis
Type: OpenClaw Skill Name: openclaw-skill-search-web Version: 0.1.0 The skill bundle is benign. It provides a web search capability using the Volcengine API, reading `VOLCENGINE_SEARCH_API_KEY` and `VOLCENGINE_SEARCH_BOT_ID` from environment variables for authentication, as documented in `SKILL.md` and `README.md`. The `scripts/volcengine_search_web.py` script makes standard API calls to `https://open.feedcoopapi.com/agent_api/agent/chat/completion` and does not contain any malicious code, unauthorized file system access, shell execution, or data exfiltration beyond its stated purpose. Instructions in `SKILL.md` are descriptive and do not attempt prompt injection against the OpenClaw agent.
Capability Assessment
Purpose & Capability
The stated purpose is web search/Q&A for OpenClaw/Trae, and the script implements that by sending the provided question to a Volcengine-style API and displaying answers, references, follow-ups, and token usage.
Instruction Scope
Instructions are explicit, user-directed CLI usage steps; there is no automatic execution, prompt override behavior, hidden workflow change, destructive action, or unrelated agent authority.
Install Mechanism
The bundle contains only README.md, SKILL.md, and one Python script. It documents the main required environment variables, though the script also accepts VOLCENGINE_ARK_API_KEY as a fallback credential.
Credentials
Using an API key, bot ID, and outbound network request is proportionate for a third-party search skill, but user queries and credentials are sent to the external API endpoint.
Persistence & Privilege
No artifact evidence shows file writes, local indexing, background workers, credential storage, privilege escalation, shell execution, self-update behavior, or persistence.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install openclaw-skill-search-web
  3. After installation, invoke the skill by name or use /openclaw-skill-search-web
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.0
volcengine-search-web skill initial release. - Provides a domestic web search solution for Openclaw/Trae using Volcengine API. - Supports web search, reference sources, multimedia cards, follow-up question suggestions, and advanced features like citations, mixed media, and glossary highlighting. - Requires environment variables for API key and bot ID. - Outputs answers with references, token usage, and error details if failed. - Optimized for stable access and Chinese-language queries without needing an international credit card.
Metadata
Slug openclaw-skill-search-web
Version 0.1.0
License
All-time Installs 51
Active Installs 51
Total Versions 1
Frequently Asked Questions

What is Search Web - CN?

Openclaw/Trae 联网搜索国内方案,使用火山引擎联网问答智能体 API 进行网络搜索问答。 It is an AI Agent Skill for Claude Code / OpenClaw, with 4150 downloads so far.

How do I install Search Web - CN?

Run "/install openclaw-skill-search-web" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Search Web - CN free?

Yes, Search Web - CN is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Search Web - CN support?

Search Web - CN is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Search Web - CN?

It is built and maintained by AlexJu (@alexjunanjing-2); the current version is v0.1.0.

💬 Comments