web-search

1# Web Search
2 
3Search the web and fetch content from URLs.
4 
5## Setup
6 
7Set `TAVILY_API_KEY` in config env to enable web search. Get your API key at https://tavily.com.
8 
9## Available Scripts
10 
11### `scripts/search.py` - Web Search
12 
13Search the web via Tavily API and return LLM-friendly markdown.
14 
15```bash
16# Search from command line
17python scripts/search.py "quantum computing" 5
18 
19# From Python
20from scripts.search import tavily_search
21output = tavily_search("quantum computing", max_results=5)
22print(output)
23```
24 
25**Parameters:**
26- `query` (str): Search query (keep under 400 chars)
27- `search_depth` (str): ultra-fast | fast | basic | advanced (default `basic`)
28- `max_results` (int): Maximum results, default 3
29- `include_raw_content` (bool): Include full page content when available
30 
31**Returns:** Markdown string with formatted results
32 
33### `scripts/fetch.py` - Fetch URL Content
34 
35Fetch and extract markdown content from a URL. No API key required.
36 
37```bash
38# Fetch from command line
39python scripts/fetch.py "https://example.com"
40 
41# From Python
42from scripts.fetch import fetch
43content = fetch("https://example.com")
44print(content)
45```
46 
47**Parameters:**
48- `url` (str): URL to fetch
49 
50**Returns:** Markdown content string
51 
52## Workflow
53 
541. **Search**: Use `tavily_search()` to find relevant pages
552. **Fetch**: Use `fetch()` to get full content from specific URLs
563. **Extract**: Parse the content to find the information you need
57 
58## Requirements
59…