Smart Search for Openclaw

Smart Search provides AI agents with a robust multi-engine web search skill featuring automatic language-based routing, seamless fallbacks, and markdown extraction capabilities.

mayf3
v1.0.0
Jun 5, 2026
0
535
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install mayf3-smart-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 mayf3-smart-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 Smart Search?

Smart Search is an advanced research and intelligence gathering utility engineered specifically for AI coding agents. It integrates multiple search backends into a unified pipeline, prioritizing a local DuckDuckGo Search (DDGS) library execution before intelligently routing and cascading requests down secondary web-fetch protocols when needed.

By incorporating this asset from Openclaw Skills, developers grant their agents the ability to dynamically navigate search filters and bypass anti-scraping blocks. The system eliminates raw HTML scraping by returning clean, structured data and provides a unified search-to-extraction mechanism that transforms targeted web links into clean Markdown text on demand.

Smart Search Use Cases

  • Automated Market Research & Fact-Checking: Efficiently query up-to-date information across various search providers.
  • Technical Intelligence Lookups: Retrieve the latest documentation, API variations, or code references during development cycles.
  • Bypassing Scraper Safeguards: Utilize robust alternative routing layers when standard search scrapers run into bot detection or blocks.
  • Multi-Lingual Information Gathering: Query seamlessly in multiple languages, ensuring contextually accurate results without manual reconfiguration.

How Smart Search Works

  1. Triggering: The AI agent triggers the search mechanism via configured command aliases or keywords (e.g., search, find info, look up).
  2. Language Detection & Routing: The system assesses the language structure of the incoming query string. CJK (Chinese, Japanese, Korean) queries are routed preferentially through Bing endpoints, while English queries default to Brave engines.
  3. Execution Layer (Tier 0): A structural request is initiated via the local DDGS engine library to extract JSON schemas containing titles, hyperlinks, and body snippets.
  4. Fallback Mechanism (Tier 1): If the Tier 0 request times out, hits empty responses, or encounters runtime errors, the pipeline falls back down an engineered matrix (e.g., routing from Brave to Yandex, Yahoo, or via web-fetch fallbacks like Bing CN, 360, and Sogou).
  5. Content Extraction: If specified by parameters, the pipeline activates the integrated extraction function to instantly fetch full-text contents from target URLs, converting them cleanly into Markdown.

Smart Search Setup

Ensure your environment is running Python 3.12. Execute the smart search utility directly from your agent's script path.

# Run a standard query with automatic language detection and fallback enabled
python3.12 {{SKILL_DIR}}/scripts/ddgs_search.py -q "Openclaw Skills" --fallback -m 5

# Check the current version of the search skill utility
python3.12 {{SKILL_DIR}}/scripts/ddgs_search.py --version

# Perform a search combined with immediate markdown content extraction
python3.12 {{SKILL_DIR}}/scripts/ddgs_search.py -q "AI research agents" --extract

Smart Search Data Schema & Taxonomy

Smart Search delivers predictable structural footprints by standardizing disparate engine outputs into organized schemas.

Primary Response Format (JSON Option)

Property Type Description
title String The header title text of the indexed search index result.
href String The fully-qualified destination URL.
body String A descriptive text snippet or summary context provided by the engine.

Time-Based Filtering Parameters

Operator Parameter Duration Mapped
t d Filter by past 24 hours (Day)
t w Filter by past 7 days (Week)
t m Filter by past 30 days (Month)
t y Filter by past 365 days (Year)

Smart Search Advanced Features

  • Intelligent Language-Based Routing: Automatically executes English queries via Brave and Chinese queries via Bing to yield optimized relevancy curves.
  • Unified Extraction Engine: Leverages built-in markdown extractors to grab full target webpage content inside a ~0.5s execution window.
  • Multi-Engine Validation (Cross-Verification): Script cross-referencing capabilities that query 2 to 3 engines concurrently and intersect arrays to isolate highly credible search returns.
  • Batch Processing Routines: Support for passing a centralized text file container filled with multiple search lines to run complex automated batch analyses.
  • Thread-Safe Safeguards: Configurable hardware timeout parameters and query limits to safely insulate systemic loops from getting stuck or blacklisted.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*