skills.rest
robbyczgw-cla
robbyczgw-claCOMMUNITY
Explore all skills by this author →
Website𝕏@CamiAi__GitHub

web-search-plus-plugin-v2

Search the web across multiple providers and extract URL content through OpenClaw tools.

100|8|Updated Feb 19, 2026
Search Tags:#routing#web search#plugin#multi-provider#openclaw#url extraction#research mode
One-click install
npx skills add https://github.com/robbyczgw-cla/web-search-plus-plugin --skill web-search-plus-plugin-v2-robbyczgw-cla
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: web-search-plus-plugin-v2
Source: https://github.com/robbyczgw-cla/web-search-plus-plugin
Command: npx skills add https://github.com/robbyczgw-cla/web-search-plus-plugin --skill web-search-plus-plugin-v2-robbyczgw-cla

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Agents often lack reliable access to current web information and full page content, and single-provider search APIs can miss sources, fail silently, or return synthesized answers instead of verifiable sources. This plugin provides source-only multi-provider web search, research orchestration, and URL extraction inside OpenClaw with deterministic routing and bounded outputs. ## Core Features & Use Cases - Multi-provider web search: web_search_plus routes queries across Serper, Brave, Tavily, Exa, Linkup, You.com, and more, with class-aware Routing v2, freshness filters, locale defaults, and canonical-source reranking. - Research mode: Runs up to three providers concurrently, deduplicates results with attributed provenance, applies a quality quorum, and extracts top sources into source_summaries for grounding. - URL extraction: web_extract_plus performs Tavily-first extraction with fallback, bounded context budgets, semantic spans, and process-local full-text references. - Operations tooling: web_routing_config_plus, web_search_health_plus, and web_extract_benchmark_plus manage routing preferences, provider health, and bounded extraction benchmarks. - Use Case: Ask the agent to research a regulatory change; it searches multiple providers concurrently, deduplicates evidence, extracts the top official sources, and returns cited summaries with routing diagnostics. ## Quick Start Enable the five optional tools via tools.alsoAllow, configure at least one provider API key such as You.com or Serper in the OpenClaw plugin config, then ask the agent to search the web for a current topic using web_search_plus.

Frequently Asked Questions about web-search-plus-plugin-v2

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I enable the web search tools in OpenClaw?▼

Add the exact tool names to tools.alsoAllow in your OpenClaw configuration, such as web_search_plus and web_extract_plus. All five tools are registered as optional, so installing the plugin alone does not make them callable.

Which search providers does web_search_plus support?▼

It supports Serper, Brave, Tavily, Exa, Querit, Linkup, Firecrawl, Parallel, SerpBase, You.com, SearXNG, Keenable, Octen via Monid, TinyFish, and DonSeTch. You only need one configured provider to start; You.com, Serper, and Linkup are the recommended starter stack.

How do I extract full text from a URL with this plugin?▼

Call web_extract_plus with the target URLs; it tries providers in Tavily-first fallback order and returns bounded inline content. For long pages, use the returned full_content_ref with content_start and content_end to read additional ranges from the process-local cache.

Does web_search_plus support a research or deep search mode?▼

Yes, set mode to research to query up to three providers concurrently, deduplicate results across them, and extract top sources into source_summaries. A default quality quorum can stop waiting once enough unique results and domains are collected.

Why is a provider like DonSeTch or TinyFish not used automatically?▼

Guarded providers such as DonSeTch, TinyFish, Octen, SerpBase, and Querit default to auto_allow=false, so they only run on explicit requests. Operators can opt them into automatic routing with web_routing_config_plus using the set_auto_allow action.

Can web_extract_plus fetch private or intranet URLs?▼

No, extraction rejects private and internal targets by default, including loopback, RFC1918, and cloud metadata addresses. Trusted intranet extraction requires explicitly setting extractAllowPrivateUrls to true in the plugin config.

Related Skills

Explore More Software Engineering Skills→
CraftOS-devCraftOS-dev
375

web-search-plus

Route search queries to specialized providers based on intent and entities.

Community
Advanced
SocketDevSocketDev
125

researching-recency

Monitor developer signals across GitHub, Hacker News, Reddit, and more.

Official
Intermediate
tomwangowatomwangowa
1

narrative-auditor

Extract and verify claims from articles against primary sources.

Community
Advanced
Zhow01Zhow01
52

web-search-plus

Route user queries to suitable search providers by analyzing intent and signals.

Community
Advanced