Explore and traverse concept relationships in your Maenifold knowledge graph through intelligent graph navigation. This tool discovers connections between [[concepts]] by analyzing co-occurrence patterns in your memory files, building contextual understanding through relationship exploration and multi-hop graph traversal.
| Parameter | Type | Required | Description | Example |
|---|---|---|---|---|
| conceptName | string | Yes | CONCEPT name to build context around (NOT a file!) | ”Machine Learning” |
| depth | int | No | How many hops in the CONCEPT GRAPH (default: 2) | 3 |
| maxEntities | int | No | Max entities to return (default: 20) | 50 |
| includeContent | bool | No | Include full content (default: false) | true |
{
"conceptName": "Neural Networks"
}
Discovers direct relationships to Neural Networks concept with default 2-hop depth and 20 entity limit.
{
"conceptName": "GraphRAG",
"depth": 1,
"maxEntities": 10
}
Explores only immediate (1-hop) relationships to GraphRAG, limiting results to 10 most relevant concepts.
{
"conceptName": "Transformer Architecture",
"depth": 3,
"maxEntities": 50
}
Performs deep 3-hop traversal around Transformer Architecture, returning up to 50 related concepts for comprehensive context.
{
"conceptName": "Memory Systems",
"depth": 2,
"maxEntities": 30,
"includeContent": true
}
Builds research context around Memory Systems with 2-hop exploration, returning 30 entities with full content for detailed analysis.
Use BuildContext before starting research to understand the conceptual landscape. Discover related topics, competing approaches, and knowledge gaps in your current understanding.
Build context around key concepts before Sequential Thinking sessions to prime your analysis with related ideas and connection patterns.
Explore concept neighborhoods to identify missing relationships or underdeveloped areas in your knowledge graph.
Use multi-hop traversal to discover conceptual bridges between different domains, revealing unexpected connections in your knowledge.
Build comprehensive context before structured workflows like design thinking or problem solving to ensure rich conceptual input.
Analyze how concepts cluster together through co-occurrence patterns, revealing natural knowledge organization in your memory system.
Perfect for understanding immediate concept connections and finding most relevant related ideas.
{"conceptName": "Deep Learning", "depth": 1}
Ideal balance of breadth and focus, discovering concept neighborhoods while maintaining relevance.
{"conceptName": "Attention Mechanism", "depth": 2}
Use for discovering distant relationships and finding unexpected conceptual bridges across domains.
{"conceptName": "Optimization", "depth": 3, "maxEntities": 40}
Cause: The concept doesn’t exist in your knowledge graph or needs different spelling
Solution: Run Sync to update graph from recent memory files, or check concept spelling and capitalization
Cause: Knowledge graph is empty or outdated compared to memory files
Solution: Execute Sync tool to extract concepts from memory files and build/update the graph database
Cause: Concept exists but appears in isolation without co-occurring with other [[concepts]]
Solution: Add the concept to existing memory files alongside related concepts to build relationships
Cause: Your knowledge graph has sparse connections or concept clusters are isolated
Solution: Increase maxEntities parameter or create memory files that bridge concept areas with shared [[concepts]]
Cause: Very dense concept graphs with high connectivity can create large result sets
Solution: Reduce depth to 1-2 or lower maxEntities to 20-30 for faster traversal
Cause: Related concepts may exist in different memory files without co-occurrence
Solution: Create bridging memory files that mention related concepts together to establish graph connections
BuildContext implements LazyGraphRAG principles through:
Concepts that appear together in the same memory file create weighted edges based on frequency of co-occurrence across multiple files.
Explores concept relationships level by level, prioritizing closer relationships while extending to discover broader context.
Relationships are ranked by co-occurrence count, ensuring most frequent concept pairings appear first in results.
Every relationship tracks which memory files contain the co-occurring concepts, enabling verification and deeper investigation.
Use BuildContext to gather conceptual context before thinking sessions:
BuildContext → SequentialThinking (with rich concept context)
Build comprehensive context before structured methodologies:
BuildContext → Design Thinking Workflow (with related concepts)
As you create more memory files with [[concepts]], BuildContext discoveries become richer and more connected.
BuildContext reveals the structure of your knowledge while SearchMemories finds the content - complementary tools for knowledge exploration.
BuildContext follows Maenifold’s Ma Protocol principles:
This tool transforms your memory files into an explorable knowledge graph, revealing the hidden connections and patterns in your accumulated knowledge through intelligent relationship traversal.