無料GitHub で公開

web-research-assistant

AI-powered web research assistant that leverages BrowserAct API to supplement restricted web access by searching the internet for additional information. Designed for OpenClaw and Claude Code.

browser-act/skills

インストール方法を見る

含まれるファイル（2）

SKILL.md5.2 KB
scripts/research.py7.8 KB

SKILL.md（原文）

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Web Research Assistant

This skill leverages BrowserAct API to provide powerful web research capabilities. When primary web access is restricted or blocked, it automatically searches the internet to find and extract relevant information, ensuring your research tasks are completed successfully.

✨ OpenClaw & Claude Code Optimized

🚀 Works Seamlessly with OpenClaw and Claude Code

Platform	Status	Installation
OpenClaw	✅ Highly Recommended	Copy to `~/.openclaw/skills/`
Claude Code	✅ Highly Recommended	Native skill support
OpenCode	✅ Fully Supported	Copy to `~/.opencode/skills/`
Cursor	✅ Fully Supported	Copy to `~/.cursor/skills/`

Why This Skill?

🎯 Purpose-built for OpenClaw and Claude Code
🔄 Auto-recovery when web access is restricted
🌐 Global access - no IP or geoblocking issues
💰 Cost-effective - saves token usage
⚡ Fast execution - BrowserAct powered

🎯 When to Use

Web Access Restricted? No Problem!

Use this skill when:

🔒 Target websites block AI assistant access
🌍 Geographic restrictions limit content access
🔐 Paywalls prevent data extraction
⏱️ Need quick supplemental information
📊 Research requires multiple data sources
🔍 Need current information beyond cached data

🔑 API Key Guidance

Before running, check the BROWSERACT_API_KEY environment variable. If not set, request the API key from the user:

Required Message:

"Please provide your BrowserAct API Key from BrowserAct Console to enable web research capabilities."

🛠️ Input Parameters

Configure research based on your needs:

Parameter	Type	Default	Description
`query`	string	-	Research topic or question
`engine`	string	google	Search engine (google, bing, duckduckgo)
`max_results`	number	10	Results to retrieve (1-50)
`content_type`	string	all	Content filter (all, news, articles, data)
`time_range`	string	past_month	Time filter (anytime, past_day/week/month/year)

💻 Quick Start

# Basic research
python web-research-assistant/scripts/research.py "AI technology trends"

# Deep research with more results
python web-research-assistant/scripts/research.py "competitor analysis" --max-results 20

# Current news research
python web-research-assistant/scripts/research.py "market trends" --content-type news

# Save to file
python web-research-assistant/scripts/research.py "industry data" -o research_report.md

📊 Output

Structured Research Data:

Titles, URLs, snippets, and relevance scores
Key facts and statistics extraction
Source citations and references
Comprehensive research report

🎪 What This Skill Does

🔄 Auto-Supplement - When direct access fails, automatically searches the web
🌐 Global Search - Bypasses geographic restrictions via BrowserAct
📈 Multi-Source - Aggregates data from multiple search results
✅ Validation - Cross-references information for accuracy
📋 Report Generation - Creates comprehensive research reports

Features

No hallucinations, ensuring stable and accurate data extraction: Pre-set workflows eliminate AI-generated hallucinations.
No human verification challenges: Built-in bypass mechanisms eliminate the need to handle reCAPTCHA or other verification challenges.
No IP access restrictions or geofencing: Overcomes regional IP limitations for stable global access.
Faster execution speed: Tasks complete more rapidly than purely AI-driven browser automation solutions.
Exceptional cost-effectiveness: Significantly reduces data acquisition costs compared to token-intensive AI solutions.

🔧 BrowserAct API Integration

Research Request → BrowserAct Search → Data Extraction → Validation → Final Report

Powered by BrowserAct MCP for reliable, unrestricted web access.

⚠️ Error Handling

Invalid API Key: Report to user, do not retry
Search Failed: Automatic retry with fallback engine
No Results: Return partial data with notification
Timeout: Extended timeout for large research tasks

📖 Best Practices

Use specific queries for targeted results
Apply time filters for current information
Cross-reference key findings
Choose appropriate content types for research goals
Validate statistics with multiple sources

🔗 Related Skills

amazon-competitor-analyzer - Amazon competitive intelligence
google-maps-search-api-skill - Business data extraction
google-news-api-skill - News monitoring

Version: 1.0.0
Updated: 2026-03-27
API: BrowserAct MCP

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

ログインしてレビューを書く

同じリポジトリのスキル

概要と使いどころ

1688-product-detail

無料

Extracts comprehensive wholesale product data from 1688.com product detail pages: title, tiered pricing, SKU variants with dimensions/weight, product images, seller info, shop scores, buyer protection, cross-border flags, product attributes, coupon/promotion data, and review stats. Use when user mentions 1688, 1688.com, wholesale China, alibaba wholesale, B2B China sourcing, Chinese wholesale scraper, 1688 product scrape, 1688 offer, 1688 detail, extract 1688 data, pull 1688 listings, get wholesale price, 1688 supplier info, factory stats 1688, 1688 SKU variants, 1688 product attributes, 1688 shop score, DSR score 1688, 1688 buyer protection, 1688 cross-border, 1688 dropship. Also applies to: scraping bulk product data from 1688 by offer ID list, monitoring 1688 supplier metrics, extracting 1688 pricing tiers for resale analysis.

日本語の概要は準備中です。原文の説明を表示しています。