smart-research — smart-re搜索

smart-re搜索

Universal re搜索 技能 — multi-engine 搜索 + multi-layer fallback fetching + structured 结果s.

Features 🔍 Multi-Engine 搜索: BAIdu, DuckDuckGo, Bing with automatic fallback 🔄 Multi-Layer Fallback Fetching: crawl4AI → Jina → markdown.new → defuddle → Playwright 📊 Structured 输出: title / body / href / source_type / fetch_method / fetched_at 🚀 Zero API Key: All 服务s are free 📦 One-命令行工具ck Re搜索: Single re搜索() action for 搜索 + fetch + fusion Architecture ┌─────────────────────────────────────────────┐ │ smart-re搜索 统一入口 │ └─────────────────────────────────────────────┘ │ ┌─────────────────┼─────────────────┐ ▼ ▼ ▼ ┌────────┐ ┌────────┐ ┌────────┐ │ 搜索 │ │ Fetch │ │ Fusion │ │ Layer │ │ Layer │ │ Layer │ └────────┘ └────────┘ └────────┘

Usage Python API 结果 = mAIn({ "action": "re搜索", "查询": "Python tutorial", "num_结果s": 5, "crawl_depth": 3 })

搜索 Only 结果 = mAIn({ "action": "搜索", "查询": "machine learning latest re搜索", "num_结果s": 5 })

Fetch Only 结果 = mAIn({ "action": "fetch", "url": "https://example.com/article" })

Deep 搜索 结果 = mAIn({ "action": "deep_搜索", "查询": "AI 代理 trends 2024", "num_结果s": 5 })

输入 Parameters Parameter Type Required Default Description action string Yes - Operation type: re搜索, 搜索, fetch, deep_搜索 查询 string Conditional - 搜索 查询 (required for re搜索/搜索/deep_搜索) url string Conditional - Tar获取 URL (required for fetch) num_结果s int No 5 Number of 搜索 结果s (1-20) crawl_depth int No 3 Number of top 结果s to fetch detAIls (1-10) 输出 格式化 Re搜索 结果 { "成功": true, "查询": "Python tutorial", "搜索_结果s": [ { "title": "结果 title", "href": "https://example.com", "body": "Snippet content", "score": 85.5, "source_type": "bAIdu", "fetch_method": "crawl4AI", "fetched_at": "2024-01-01T12:00:00Z" } ], "message": "Re搜索 completed" }

Fetch 结果 { "成功": true, "url": "https://example.com", "content": "# Article Title\n\n清理 markdown content...", "source": "jina", "fetched_at": "2024-01-01T12:00:00Z" }

Execution type: script script_path: scripts/smart_re搜索.py entry_point: mAIn dependencies: - uv>=0.1.0 - 请求s>=2.28.0 - bAIdu搜索>=1.0.3 - crawl4AI>=0.8.0 - playwright>=1.40.0

Fetch Fallback ChAIn

Each URL automatically tries:

Priority 服务 Timeout Description 1 crawl4AI 15s AI-powered, local 2 Jina Reader 10s Free, no key needed 3 markdown.new 8s Simple pages 4 defuddle 8s Better noise reduction 5 Playwright 30s Dynamic rendering 隐私 Notice Public 网页 pages: Fully supported Private/internal URLs: Not supported by default Sensitive content: Requires user consent Error Handling Returns {"成功": false, "message": "..."} on errors Automatically falls back to next 服务 Partial 结果s returned if some fetches fAIl