WriteMemory

Create new knowledge files in the Maenifold memory system with automatic knowledge graph integration. This tool builds your persistent knowledge base through [[WikiLink]] concepts that automatically connect related information. REQUIRES at least one [[concept]] in double brackets to ensure knowledge graph connectivity.

When to Use This Tool

Creating new knowledge files with structured information
Capturing insights, research, or analysis that should persist across sessions
Starting new knowledge areas or topics with proper graph connections
Documenting solutions, patterns, or discoveries for future reference
Building interconnected knowledge networks through concept relationships
Storing structured information that will be searched or referenced later
Creating memory files that integrate with Sequential Thinking or Workflow sessions

Key Features

Automatic Knowledge Graph Integration: Every [[concept]] becomes a searchable graph node
WikiLink Enforcement: Validates content contains at least one [[concept]] for connectivity
Flexible Organization: Optional folder structure for logical grouping
Tagging System: Add categorical tags for enhanced discoverability
Ma Protocol Compliance: Files stay under 250 lines, encouraging focused content
Automatic Metadata: Generates timestamps, permalinks, and file structure
Checksum Generation: Provides content verification for safe concurrent editing
URI-based Access: Creates memory:// URIs for consistent file references

Parameters

Parameter	Type	Required	Description	Example
title	string	Yes	Human-readable title for the memory file	”Machine Learning Fundamentals”
content	string	Yes	Main content with required [[concepts]] in brackets	”Learning about [[Neural Networks]] and [[Deep Learning]] applications in [[Computer Vision]]“
folder	string	No	Optional folder path for organization	”research/ai” or “projects/2024”
tags	string[]	No	Optional tags for categorization	[“ai”, “research”, “fundamentals”]

Usage Examples

Basic Knowledge Creation

{
  "title": "GraphRAG Implementation Notes",
  "content": "Implementing [[GraphRAG]] requires understanding [[Vector Embeddings]] and [[Knowledge Graphs]]. Key insight: combine [[Semantic Search]] with [[Graph Traversal]] for enhanced context retrieval."
}

Creates a basic memory file with automatic concept extraction and graph integration.

Organized Research File

{
  "title": "Transformer Architecture Deep Dive",
  "content": "The [[Transformer]] architecture revolutionized [[Natural Language Processing]] through [[Self-Attention]] mechanisms. Key components include [[Multi-Head Attention]], [[Positional Encoding]], and [[Layer Normalization]].\n\n## Key Insights\n- [[Attention]] allows parallel processing unlike [[RNN]] sequential approach\n- [[BERT]] and [[GPT]] are both [[Transformer]]-based but use different training objectives",
  "folder": "research/deep-learning",
  "tags": ["transformers", "nlp", "architecture", "research"]
}

Creates an organized research file with folder structure and comprehensive tagging.

Problem-Solution Documentation

{
  "title": "Debugging Memory Leaks in C# Applications", 
  "content": "Encountered [[Memory Leak]] in [[C# Application]] during [[Performance Testing]]. Root cause: [[Event Handlers]] not properly unsubscribed in [[Observer Pattern]] implementation.\n\n## Solution\n- Use [[Weak References]] for event subscriptions\n- Implement [[IDisposable]] pattern properly\n- Apply [[RAII]] principles from [[Resource Management]]",
  "folder": "solutions/performance",
  "tags": ["debugging", "performance", "csharp", "memory-management"]
}

Documents a specific problem-solution pattern with technical concepts.

Common Patterns

Research and Learning Notes

Use WriteMemory to capture learning insights with concept connections. Every new concept becomes searchable and linkable to future knowledge.

Project Documentation

Create project-specific memory files in dedicated folders. Use consistent tagging schemes for project phases or components.

Solution Cataloging

Document solutions to technical problems with [[concept]] links to technologies, patterns, and methodologies used.

Sequential Thinking Integration

WriteMemory works seamlessly with Sequential Thinking sessions - capture thinking results as persistent knowledge for future reference.

Knowledge Graph Building

Each [[concept]] becomes a node. Related files sharing concepts automatically connect, building your knowledge graph organically.

SearchMemories: Find existing knowledge by content or concepts before creating new files
ReadMemory: Access created files by URI or title for reference or editing
EditMemory: Modify existing files while maintaining [[concept]] requirements
Sync: Rebuild knowledge graph after creating multiple files
BuildContext: Explore concept relationships from files you’ve created
SequentialThinking: Reference memory files in thinking sessions and capture results back to memory

Folder Organization Patterns

Recommended Folder Structure

research/           # Research notes and findings
projects/           # Project-specific documentation  
solutions/          # Problem-solution pairs
methodologies/      # Process and methodology documentation
chain-of-thought/   # Automatic from thinking sessions
concepts/           # Pure concept definitions

Folder Best Practices

Use hierarchical paths like research/ai/transformers for deep organization
Keep folder names simple and descriptive
Consider using dates in folders like projects/2024/q4 for temporal organization
Match folder structure to your workflow and thinking patterns

Troubleshooting

Error: “Content must contain at least one [[concept]]”

Cause: Content lacks [[WikiLink]] concepts in double brackets
Solution: Add relevant concepts like [[Machine Learning]] or [[Software Architecture]] to connect your knowledge to the graph

Error: “Memory file not found during read”

Cause: File title or URI reference is incorrect
Solution: Use exact title from WriteMemory response or full memory:// URI format

Warning: Creating orphaned knowledge

Cause: Using very unique [[concepts]] that don’t connect to existing knowledge
Solution: Include some established concepts alongside new ones to maintain graph connectivity

File organization issues

Cause: Inconsistent folder naming or deep nesting
Solution: Use consistent folder patterns and avoid more than 3-4 levels of nesting

Content too long (Ma Protocol violation)

Cause: Single file exceeds 250 lines
Solution: Split into multiple focused files with shared [[concepts]] for connection

Knowledge Graph Integration

Every WriteMemory operation automatically:

Extracts [[concepts]] from content using WikiLink parsing
Creates graph nodes for new concepts or connects to existing ones
Builds relationships between concepts within the same file
Enables traversal through BuildContext and Visualize tools
Powers search through concept-based queries in SearchMemories

The [[concept]] requirement ensures no orphaned knowledge - every file connects to your growing knowledge graph, making information discoverable and relationship-aware.

Ma Protocol Compliance

WriteMemory follows Maenifold’s Ma Protocol principles:

Simplicity: Single responsibility for knowledge creation
No Magic: Real file system storage with transparent metadata
Minimal Complexity: Static methods, clear parameters, direct operation
Real Testing: End-to-end file creation and validation
250-line Limit: Encourages focused, coherent knowledge units

This tool creates the foundation for your Maenifold knowledge system - every file becomes part of your growing cognitive infrastructure.