Exa Search Integration
Quick Reference
| Topic |
When to Use |
Reference |
| Search Modes |
Choosing between auto, neural, and keyword search |
search-modes.md |
| Filters |
Domain, date, text, and category filtering |
filters.md |
| Contents |
Text extraction, highlights, summaries, livecrawl |
contents.md |
| SDK Patterns |
Python (exa_py) and TypeScript (exa-js) usage |
sdk-patterns.md |
Essential Patterns
Basic Search (Python)
from exa_py import Exa
exa = Exa(api_key="your-api-key") # or set EXA_API_KEY env var
results = exa.search_and_contents(
"latest developments in quantum computing",
type="auto",
num_results=10,
text=True,
highlights=True
)
for result in results.results:
print(f"{result.title}: {result.url}")
print(result.text[:500])
Basic Search (TypeScript)
import Exa from "exa-js";
const exa = new Exa(process.env.EXA_API_KEY);
const results = await exa.searchAndContents(
"latest developments in quantum computing",
{
type: "auto",
numResults: 10,
text: true,
highlights: true,
}
);
results.results.forEach((result) => {
console.log(`${result.title}: ${result.url}`);
});
Search with Filters
results = exa.search_and_contents(
"AI startup funding rounds",
type="neural",
num_results=10,
include_domains=["techcrunch.com", "venturebeat.com"],
start_published_date="2024-01-01",
text={"max_characters": 2000},
summary=True
)
Find Similar Links
similar = exa.find_similar_and_contents(
"https://example.com/interesting-article",
num_results=10,
exclude_source_domain=True,
text=True
)
Search Mode Selection
| Mode |
When to Use |
Notes |
auto |
Default for most queries |
Exa optimizes between neural/keyword automatically |
neural |
Natural language, conceptual queries |
Best for "what is...", "how to...", topic exploration |
keyword |
Exact matches, technical terms, names |
Best for specific product names, error codes, proper nouns |
Common Mistakes
- Using
keyword for conceptual queries - Neural search understands intent better; use auto or neural for natural language questions
- Not setting
text=True - Search returns URLs only by default; explicitly request content with text=True
- Ignoring
highlights - Use highlights=True for relevant snippets without downloading full page text
- Missing API key - Set
EXA_API_KEY environment variable or pass explicitly to constructor
- Over-filtering initially - Start with broad searches, then add domain/date filters to refine
- Not using
summary - For RAG applications, summary=True provides concise context without full page text
- Expecting scores in auto mode - Relevance scores are only returned with
type="neural"; auto mode doesn't include them
1---2name: exa-search3description: Exa.ai search API integration for neural and keyword web search with content retrieval. Use when implementing web search features, integrating Exa SDK (exa_py, exa-js), or retrieving web content. Triggers on: Exa, exa_py, exa-js, neural search, web search API, search_and_contents, searchAndContents, find_similar, findSimilar, domain filtering, date filtering, text extraction, page summaries, highlights, search auto mode, fast search, search categories, livecrawl, excluding domains, include text, exclude text, EXA_API_KEY.4license: MIT5---67# Exa Search Integration89## Quick Reference1011| Topic | When to Use | Reference |12|-------|-------------|-----------|13| **Search Modes** | Choosing between auto, neural, and keyword search | [search-modes.md](references/search-modes.md) |14| **Filters** | Domain, date, text, and category filtering | [filters.md](references/filters.md) |15| **Contents** | Text extraction, highlights, summaries, livecrawl | [contents.md](references/contents.md) |16| **SDK Patterns** | Python (exa_py) and TypeScript (exa-js) usage | [sdk-patterns.md](references/sdk-patterns.md) |1718## Essential Patterns1920### Basic Search (Python)2122```python23from exa_py import Exa2425exa = Exa(api_key="your-api-key") # or set EXA_API_KEY env var2627results = exa.search_and_contents(28 "latest developments in quantum computing",29 type="auto",30 num_results=10,31 text=True,32 highlights=True33)3435for result in results.results:36 print(f"{result.title}: {result.url}")37 print(result.text[:500])38```3940### Basic Search (TypeScript)4142```typescript43import Exa from "exa-js";4445const exa = new Exa(process.env.EXA_API_KEY);4647const results = await exa.searchAndContents(48 "latest developments in quantum computing",49 {50 type: "auto",51 numResults: 10,52 text: true,53 highlights: true,54 }55);5657results.results.forEach((result) => {58 console.log(`${result.title}: ${result.url}`);59});60```6162### Search with Filters6364```python65results = exa.search_and_contents(66 "AI startup funding rounds",67 type="neural",68 num_results=10,69 include_domains=["techcrunch.com", "venturebeat.com"],70 start_published_date="2024-01-01",71 text={"max_characters": 2000},72 summary=True73)74```7576### Find Similar Links7778```python79similar = exa.find_similar_and_contents(80 "https://example.com/interesting-article",81 num_results=10,82 exclude_source_domain=True,83 text=True84)85```8687## Search Mode Selection8889| Mode | When to Use | Notes |90|------|-------------|-------|91| `auto` | Default for most queries | Exa optimizes between neural/keyword automatically |92| `neural` | Natural language, conceptual queries | Best for "what is...", "how to...", topic exploration |93| `keyword` | Exact matches, technical terms, names | Best for specific product names, error codes, proper nouns |9495## Common Mistakes96971. **Using `keyword` for conceptual queries** - Neural search understands intent better; use `auto` or `neural` for natural language questions982. **Not setting `text=True`** - Search returns URLs only by default; explicitly request content with `text=True`993. **Ignoring `highlights`** - Use `highlights=True` for relevant snippets without downloading full page text1004. **Missing API key** - Set `EXA_API_KEY` environment variable or pass explicitly to constructor1015. **Over-filtering initially** - Start with broad searches, then add domain/date filters to refine1026. **Not using `summary`** - For RAG applications, `summary=True` provides concise context without full page text1037. **Expecting scores in auto mode** - Relevance scores are only returned with `type="neural"`; auto mode doesn't include them
ejirocodes