A zero-API, localized web search skill that scrapes DuckDuckGo, Bing, and Google directly with automatic engine fallback.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install local-search
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install local-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 Local Search?
Local Search is a robust, serverless utility built to give AI agents direct access to real-time information on the web without relying on external cloud APIs or software development kits. Running entirely on your local machine, the skill executes direct HTTP requests to public search engines and parses the returned HTML structure using Cheerio. This minimizes overhead, eliminates subscription costs, and provides absolute privacy for queries.
As a highly versatile option among Openclaw Skills, this tool implements an intelligent auto-fallback orchestrator. If one search provider issues a rate limit, the system transparently falls back to secondary and tertiary providers in a DuckDuckGo to Bing to Google chain. This guarantees that your local workflows and autonomous agents always receive structured web information, even when running on typical datacenter or cloud environments where conventional scraping frequently fails.
Local Search Use Cases
- Real-time Context Gathering: Retrieve live news, current stock performance, or up-to-date weather data for immediate prompt context.
- Fresh Tech Documentation: Fetch recently published framework release notes, software API schemas, or code repositories beyond the model's knowledge cutoff.
- Unbiased Fact-Checking: Verify raw claims against immediate live web results to prevent model hallucination.
- Competitor and Market Analysis: Track ongoing corporate press releases, product updates, and industry changes dynamically.
- Aggregated Academic Research: Query across engines to build bibliographies, verify paper citations, or extract authors from public repositories.
How Local Search Works
- Request Reception: The local-search skill accepts a query along with configuration parameters such as result limit, engine choice, locale, or recency restrictions.
- Engine Sequence Selection: If configured in "auto" mode, the orchestrator sets up an internal chain starting with DuckDuckGo, proceeding to Bing, and ending with Google.
- Direct HTML Fetch: The skill performs standard HTTP requests (POST or GET) using Node's native fetch capabilities, mimicking standard client request behaviors.
- Robust Parsing: Cheerio inspects raw HTML elements and extracts essential DOM segments including headings, anchor tags, snippets, and favicons.
- Schema Harmonization: Results are mapped directly into a standardized structure, providing uniform compatibility with other Openclaw Skills.
- Auto-Fallback Trigger: If the chosen engine returns a soft failure, rate limit, or captcha redirect, the runner catches the exception and immediately falls back to the next engine in sequence.
Local Search Setup
Get started immediately by installing the skill directly to your workspace:
npx clawhub install @Sakurakilove/local-search
Alternatively, you can manually clone the skill into your designated projects directory:
# Navigate to your local-search folder
cd skills/local-search
# Install the direct dependency (cheerio)
npm install
# Execute a quick test query via tsx
tsx bin/web-search.ts "artificial intelligence"
Local Search Data Schema & Taxonomy
The local-search skill structures all output using the canonical SearchFunctionResultItem format, allowing seamless interoperability with other Openclaw Skills.
Output Fields Schema
| Field Name | Type | Description |
|---|---|---|
url |
string | Full target destination URL of the search result. |
name |
string | Page title of the scraped search result. |
snippet |
string | Short context preview text extracted from the page. |
host_name |
string | Parsed host domain (e.g., "wikipedia.org"). |
rank |
number | 1-indexed relative rank on the parsed results page. |
date |
string | Extracted publishing or modification date ("N/A" if unknown). |
favicon |
string | Google S2-cached Favicon URL of the target domain. |
source_engine |
string | The specific engine that successfully resolved the query. |
raw_html |
string (optional) | Original snippet DOM HTML extracted by Cheerio. |
score |
number (optional) | Relative relevance rating score on a scale from 0 to 100. |
Local Search Advanced Features
- Locale Configuration: Explicitly target localized engine indexes via standard BCP-47 tags (e.g.,
--locale ja-JP) to avoid regional parsing biases. - Strict Recency Constraints: Restrict search scopes to specified timespans using custom recency flags (e.g.,
--recency-days 7). - Pure Programmatic SDK: Easily import the orchestrator's core functional logic directly into standard TypeScript or JavaScript code environments.
- Custom Fetch Overrides: Pass specialized mock handlers or user-agent overrides directly into runtime configurations for testing purposes.
- Scriptable Quiet Output: Silence default runtime logging info with quiet flags, enabling easy piping of clean JSON search outputs.
SKILL.md
Loading
Related Openclaw Skills
Web3 & Open Banking Finance Administration Skills
A professional operational framework and skill designed to integrate on-chain Web3 treasury management with traditional Open Banking API systems and compliance automation workflows.
slim
A pluggable pipe filter that strips verbose CLI output before it reaches the LLM to maximize your context budget.
Sharebench Artifact Search
Search Sharebench, an open registry of reusable AI artifacts like skills, personas, prompts, and playbooks in SKILL.md format to avoid building from scratch.
Mailbuttons Governed Agent Email Setup
Mailbuttons enables developers to securely assign governed email inboxes to AI agents, ensuring complete sandbox safety and human-in-the-loop validation.
Public IP Watch
A lightweight monitoring skill to track your public IP address and notify you instantly when changes occur, preventing DNS and connection breaks.
E-commerce Video Script Generator
An automated e-commerce video script generator that translates product selling points into highly localized 10-second and 15-second multi-language storyboard scripts.
Featured*
