agent-websearch-skill
Intelligent multi-engine web search with automatic switching, no VPN required.
Install & Use
Copy this prompt and send it to your AI assistant (Claude / Cursor / TRAE / Codex / WorkBuddy etc.) to auto-install:
Help me install this AI Skill: agent-websearch-skill. It is used for: Intelligent multi-engine web search with automatic switching, no VPN required. Full Skill content: https://321skill.com/skills/agent-websearch-skill/raw/index.md Read that page and install it.
The prompt includes a link to the full Skill content. You can also view the full content.
This Skill addresses issues such as search engine unavailability, insufficient API quotas, and variable network conditions during web searches. In practical development, developers often need to quickly obtain the latest information but may encounter awkward situations like DuckDuckGo being blocked, Tavily quotas being exhausted, or Bing API not being configured. Manually switching engines is both cumbersome and inefficient.
Usage is straightforward: after installing the dependencies, simply call the search() function with your query keywords and optional parameters. You can then trigger searches via natural language commands, such as "Search for the latest Python tutorial." It automatically detects the current network environment, selects an available engine based on priority, and returns structured results. It also supports advanced features like full webpage scraping, batch searches, and viewing system status.
It is well-suited for teams or individuals who frequently conduct web searches. This is particularly true for developers who have already set up AI Agent workflows and need to automatically fetch external information to aid decision-making, as well as for academic researchers, content creators, and other users who require real-time web data.
It is recommended to set prefer_quality=True to prioritize using Tavily when high-quality search results are needed. Use the default balanced mode for free-first scenarios. Note that DuckDuckGo may still be inaccessible in some network environments; in such cases, the skill will automatically fall back to the Bing crawler. However, the Bing crawler may be affected by anti-crawling policies, so it is advisable to configure Tavily or Bing API as a backup.
Key Features
Unlike single-engine search tools (e.g., using duckduckgo_search directly), it automatically detects the network environment and intelligently switches engines, eliminating the need for manual VPN or API key configuration, while also featuring built-in quota management to avoid exceeding call limits.
Limitations
Requires Python 3.8+ environment and installation of three libraries: requests, tavily, and duckduckgo_search. Some engines (e.g., DuckDuckGo) may still be inaccessible in certain network environments, and the Bing crawler mode is susceptible to anti-crawling policies.
FAQ
Does this skill require an API Key?
No, it defaults to using DuckDuckGo (free and unlimited). However, if you wish to use Tavily or Bing API for higher quality results, you need to configure the corresponding API Key.
How do I force an engine switch?
Set `prefer_quality=True` to prioritize using Tavily, or set `force_network_check=True` to force a re-detection of the network environment and re-select an engine.
Installation guide for AI assistants
If your AI coding assistant (Claude Code, Cursor, TRAE etc.) can see this page, send it this message to auto-install:
Visit https://321skill.com/skills/agent-websearch-skill/raw/index.md to read the original Skill definition (Markdown format) for agent-websearch-skill, and install it according to the instructions.
Raw Markdown URL for AI: /skills/agent-websearch-skill/raw/index.md