Search Examples | Airweave

This page provides quick examples to get you started with Airweave search. For a comprehensive, hands-on tutorial with visualizations and real-world scenarios, check out our interactive Jupyter notebook.

Interactive Tutorial

Learn search concepts through hands-on examples in our comprehensive Jupyter notebook with live code, visualizations, and real-world scenarios.

View the Advanced Search Tutorial →

Quick Start

1 from airweave import AirweaveSDK
2 from airweave.schemas.search import SearchRequest
3 from qdrant_client.http.models import Filter, FieldCondition, MatchValue
4 
5 client = AirweaveSDK(api_key="your-api-key")

Essential Examples

Basic Search

1 # Simple text search
2 response = await client.collections.search_collection(
3     readable_id="your-collection-id",
4     query="customer onboarding process"
5 )

Search with AI Completion

1 # Get AI-generated insights
2 response = await client.collections.search_collection(
3     readable_id="your-collection-id",
4     query="What are our security policies?",
5     response_type="completion"
6 )

Filtered Search

1 # Search within specific source (⚠️ case-sensitive!)
2 request = SearchRequest(
3     query="API documentation",
4     filter=Filter(
5         must=[
6             FieldCondition(
7                 key="source_name",
8                 match=MatchValue(value="GitHub")  # Must match exactly
9             )
10         ]
11     )
12 )
13 
14 response = await client.collections.search_collection_advanced(
15     readable_id="your-collection-id",
16     search_request=request
17 )

Date Range Filter

1 from datetime import datetime, timezone, timedelta
2 from qdrant_client.http.models import DatetimeRange
3 
4 # Find items from last 7 days
5 request = SearchRequest(
6     query="bug fixes",
7     filter=Filter(
8         must=[
9             FieldCondition(
10                 key="created_at",
11                 range=DatetimeRange(
12                     gte=datetime.now(timezone.utc) - timedelta(days=7)
13                 )
14             )
15         ]
16     )
17 )

Case-Insensitive Source Matching

1 from qdrant_client.http.models import MatchAny
2 
3 # Handle different case variations
4 request = SearchRequest(
5     query="deployment guide",
6     filter=Filter(
7         must=[
8             FieldCondition(
9                 key="source_name",
10                 match=MatchAny(any=["GitHub", "github", "GITHUB"])
11             )
12         ]
13     )
14 )

REST API

Try the advanced search endpoint with filters in our interactive API playground. Open API Explorer →

$ # GET - Basic search
> curl -X GET "https://api.airweave.ai/api/v1/collections/{id}/search?query=test" \
>   -H "x-api-key: your-api-key"
> 
> # POST - Advanced search with filters
> curl -X POST "https://api.airweave.ai/api/v1/collections/{id}/search" \
>   -H "x-api-key: your-api-key" \
>   -H "Content-Type: application/json" \
>   -d '{
>     "query": "security vulnerabilities",
>     "filter": {
>       "must": [{
>         "key": "source_name",
>         "match": {"value": "GitHub"}
>       }]
>     },
>     "score_threshold": 0.7
>   }'

Filter Patterns

This is a hypothetical example of what you might want to do.

1 # Multi-source search
2 FieldCondition(
3     key="source_name",
4     match=MatchAny(any=["Asana", "Jira", "Linear"])
5 )
6 
7 # Priority filtering
8 FieldCondition(
9     key="metadata.priority",
10     match=MatchAny(any=["high", "critical", "P0"])
11 )
12 
13 # Exclude resolved items
14 Filter(
15     must_not=[
16         FieldCondition(
17             key="metadata.status",
18             match=MatchAny(any=["closed", "resolved", "done"])
19         )
20     ]
21 )

Learn More

Search Concepts - Understand parameters and options
Using Filters - Master Qdrant filtering
API Reference - Complete API details
Interactive Tutorial ↗ - Hands-on Jupyter notebook with visualizations