Search Types Reference

Cognee provides a variety of search types to query your knowledge graph in different ways. This reference guide explains each search type, its purpose, and how to use it effectively.

Available Search Types

Cognee offers the following search types through the SearchType enum:

from cognee.modules.search.types import SearchType

# Available search types
SearchType.SUMMARIES
SearchType.INSIGHTS
SearchType.CHUNKS
SearchType.COMPLETION
SearchType.GRAPH_COMPLETION
SearchType.GRAPH_SUMMARY_COMPLETION
SearchType.GRAPH_COMPLETION_COT
SearchType.GRAPH_COMPLETION_CONTEXT_EXTENSION
SearchType.CODE
SearchType.CYPHER

Basic Search Syntax

All search types use the same basic function call pattern:

import cognee
from cognee.modules.search.types import SearchType

results = await cognee.search(
    query_text="Your search query",
    query_type=SearchType.INSIGHTS,  # Replace with your desired search type
)

Detailed Search Type Descriptions

SUMMARIES

The SUMMARIES search type retrieves summarized information from your knowledge graph.

results = await cognee.search(
    query_text="Summarize what is known about quantum computing",
    query_type=SearchType.SUMMARIES
)

Best for:

Getting concise overviews of topics
Summarizing large amounts of information
Quick understanding of complex subjects

INSIGHTS

The INSIGHTS search type discovers connections and relationships between entities in your knowledge graph.

results = await cognee.search(
    query_text="What is the relationship between Einstein and Germany?",
    query_type=SearchType.INSIGHTS
)

Best for:

Discovering how entities are connected
Understanding relationships between concepts
Exploring the structure of your knowledge graph

CHUNKS

The CHUNKS search type retrieves specific facts and information chunks from your knowledge graph.

results = await cognee.search(
    query_text="When was Einstein born?",
    query_type=SearchType.CHUNKS
)

Best for:

Finding specific facts
Getting direct answers to questions
Retrieving precise information

COMPLETION

The COMPLETION search type uses retrieval-augmented generation to provide comprehensive answers based on your knowledge graph.

results = await cognee.search(
    query_text="Explain Einstein's contributions to physics",
    query_type=SearchType.COMPLETION
)

Best for:

Getting detailed explanations
Combining multiple pieces of information
Generating comprehensive answers

GRAPH_COMPLETION

The GRAPH_COMPLETION search type leverages the graph structure to provide more contextually aware completions.

results = await cognee.search(
    query_text="How do quantum computers differ from classical computers?",
    query_type=SearchType.GRAPH_COMPLETION
)

Best for:

Complex queries requiring graph traversal
Questions that benefit from understanding relationships
Queries where context from connected entities matters

GRAPH_SUMMARY_COMPLETION

The GRAPH_SUMMARY_COMPLETION search type combines graph traversal with summarization to provide concise but comprehensive answers.

results = await cognee.search(
    query_text="Summarize the key developments in AI from 2010 to 2023",
    query_type=SearchType.GRAPH_SUMMARY_COMPLETION
)

Best for:

Getting summarized information that requires understanding relationships
Complex topics that need concise explanations
Queries that benefit from both graph structure and summarization

GRAPH_COMPLETION_COT

The GRAPH_COMPLETION_COT search type combines graph traversal with chain of thought to provide answers to complex multi hop questions.

results = await cognee.search(
    query_text="Who was the father of the person who invented the lightbulb",
    query_type=SearchType.GRAPH_COMPLETION_COT
)

Best for:

Multi-hop questions that require following several linked concepts or entities
Tracing relational paths in a knowledge graph while also getting clear step-by-step reasoning
Summarizing completx linkages into a concise, human-readable answer once all hops have been explored

GRAPH_COMPLETION_CONTEXT_EXTENSION

The GRAPH_COMPLETION_CONTEXT_EXTENSION search type combines graph traversal with multi-round context extension.

results = await cognee.search(
    query_text="What county was XY born in",
    query_type=SearchType.GRAPH_COMPLETION_CONTEXT_EXTENSION
)

Best for:

Iterative, multi-hop queries where intermediate facts aren’t all present upfront
Complex linkages that benefit from multi-round “search → extend context → reason” loops to uncover deep connections.
Sparse or evolving graphs that require on-the-fly expansion—issuing follow-up searches to discover missing nodes or properties

CODE

The CODE search type is specialized for retrieving and understanding code-related information from your knowledge graph.

results = await cognee.search(
    query_text="How to implement a binary search tree in Python",
    query_type=SearchType.CODE
)

Best for:

Code-related queries
Programming examples and patterns
Technical documentation searches

CYPHER

The CYPHER search type allows you to execute raw Cypher queries directly against your graph database.

# Find all entities and their types
results = await cognee.search(
    query_text="MATCH (n) RETURN labels(n) as types, n.name as name LIMIT 10",
    query_type=SearchType.CYPHER
)

Best for:

Executing precise graph queries with full control
Leveraging Cypher features and functions
Getting raw data directly from the graph database

Important Notes:

Only supported with Neo4j, Memgraph, and FalkorDB (not NetworkX)
Requires knowledge of Cypher query language
Returns raw query results without additional processing

Advanced Usage

Combining Search Types

For complex queries, you might want to use multiple search types in sequence:

# First get insights about relationships
insights = await cognee.search(
    query_text="Relationship between quantum computing and cryptography",
    query_type=SearchType.INSIGHTS
)

# Then get a detailed explanation
explanation = await cognee.search(
    query_text=f"Explain how quantum computing affects cryptography, considering: {insights}",
    query_type=SearchType.GRAPH_COMPLETION
)

Search with Filters

You can filter search results by node type or other properties:

# Search only for people and organizations
results = await cognee.search(
    query_text="Who founded Microsoft?",
    query_type=SearchType.INSIGHTS,
    node_types=["Person", "Organization"]
)

Choosing the Right Search Type

Here’s a quick guide to help you choose the most appropriate search type:

If you want to…	Use this search type
Get a quick summary	`SUMMARIES`
Discover relationships	`INSIGHTS`
Find specific facts	`CHUNKS`
Get detailed explanations	`COMPLETION`
Understand complex relationships	`GRAPH_COMPLETION`
Get concise answers about complex topics	`GRAPH_SUMMARY_COMPLETION`
Answers complex multi-hop questions	`GRAPH_COMPLETION_COT`
Extend context iteratively to answer complex questions	`GRAPH_COMPLETION_CONTEXT_EXTENSION`
Find code examples or technical information	`CODE`
Execute raw Cypher queries against the graph database	`CYPHER`

Examples

Example 1: Research Paper Analysis

# Add research papers to your knowledge graph
await cognee.add("file:///path/to/research_paper1.pdf")
await cognee.add("file:///path/to/research_paper2.pdf")

# Process the content
await cognee.cognify()

# Get a summary of the key findings
summaries = await cognee.search(
    query_text="Summarize the key findings from these research papers",
    query_type=SearchType.SUMMARIES
)

# Discover relationships between concepts
insights = await cognee.search(
    query_text="What is the relationship between the methodologies used in these papers?",
    query_type=SearchType.INSIGHTS
)

# Get a comprehensive analysis
analysis = await cognee.search(
    query_text="Provide a comprehensive analysis of how these papers contribute to the field",
    query_type=SearchType.GRAPH_COMPLETION
)

Example 2: Code Documentation

# Add code documentation to your knowledge graph
await cognee.add("file:///path/to/codebase")

# Process the content
await cognee.cognify()

# Find specific code examples
code_examples = await cognee.search(
    query_text="How to implement authentication in this codebase",
    query_type=SearchType.CODE
)

# Understand the architecture
architecture = await cognee.search(
    query_text="Explain the overall architecture of this codebase",
    query_type=SearchType.GRAPH_SUMMARY_COMPLETION
)

Troubleshooting

If you’re not getting the expected search results:

No results: Make sure you’ve added relevant content and run cognee.cognify()
Irrelevant results: Try refining your query or using a different search type
Missing connections: Your knowledge graph might not have captured the relationships you’re looking for; try adding more relevant content

Next Steps

Learn how to visualize your knowledge graph
Explore Cognee Tasks for specific use cases
See how to build custom pipelines for specialized search functionality

Coming Soon

​Search Types Reference

​Available Search Types

​Basic Search Syntax

​Detailed Search Type Descriptions

​SUMMARIES

​INSIGHTS

​CHUNKS

​COMPLETION

​GRAPH_COMPLETION

​GRAPH_SUMMARY_COMPLETION

​GRAPH_COMPLETION_COT

​GRAPH_COMPLETION_CONTEXT_EXTENSION

​CODE

​CYPHER

​Advanced Usage

​Combining Search Types

​Search with Filters

​Choosing the Right Search Type

​Examples

​Example 1: Research Paper Analysis

​Example 2: Code Documentation

​Troubleshooting

​Next Steps

Search Types Reference

Available Search Types

Basic Search Syntax

Detailed Search Type Descriptions

SUMMARIES

INSIGHTS

CHUNKS

COMPLETION

GRAPH_COMPLETION

GRAPH_SUMMARY_COMPLETION

GRAPH_COMPLETION_COT

GRAPH_COMPLETION_CONTEXT_EXTENSION

CODE

CYPHER

Advanced Usage

Combining Search Types

Search with Filters

Choosing the Right Search Type

Examples

Example 1: Research Paper Analysis

Example 2: Code Documentation

Troubleshooting

Next Steps