memorecall 0.1.0

A memory and recall enhancement library for AI agents.

memorecall

A memory and recall enhancement library for Python. This package provides functionality similar to the memori project for storing and retrieving memories with enhanced search capabilities.

Note: This is currently a placeholder package. Full functionality will be implemented in future versions.

Architecture Overview

The workflow diagram above illustrates the complete memory processing pipeline:

1. Ingestion: Raw data is ingested and preprocessed
2. Embedding: Content is converted to vector embeddings for semantic search
3. Indexing: Memories are indexed for efficient retrieval
4. Storage: Processed memories are stored in the configured backend
5. Search: Queries are processed and matched against stored memories
6. Retrieval: Relevant memories are returned with ranking

Installation

pip install memorecall

Quick Start

from memorecall import MemoryRecall

# Initialize the memory system
memory = MemoryRecall()

# Store memories
memory.store("Important meeting notes", tags=["work", "meeting"])
memory.store("Personal reminder", tags=["personal"])

# Recall memories
all_memories = memory.recall()
work_memories = memory.recall(tags=["work"])
search_results = memory.recall(query="meeting")

print(f"Total memories: {memory.count()}")

Features

• Memory Storage: Store text-based memories with optional tags
• Smart Recall: Search memories by content or filter by tags
• Simple API: Easy-to-use interface for memory operations
• Lightweight: Minimal dependencies

API Reference

MemoryRecall Class

__init__(config=None)

Initialize a new MemoryRecall instance.

store(memory, tags=None)

Store a memory with optional tags.

• memory (str): The content to store
• tags (list, optional): List of tags to associate with the memory
• Returns: bool - True if successful

recall(query=None, tags=None)

Retrieve memories based on search criteria.

• query (str, optional): Search query to filter by content
• tags (list, optional): List of tags to filter by
• Returns: list - List of matching memories

clear()

Clear all stored memories.

• Returns: bool - True if successful

count()

Get the number of stored memories.

• Returns: int - Number of memories

Development

This is a placeholder package. The current implementation provides basic in-memory storage and retrieval. Future versions will include:

• Persistent storage options
• Advanced search algorithms
• Memory categorization and organization
• Export/import functionality
• Integration with external memory systems

Contributing

This project is currently in placeholder status. Contributions will be welcome once the main implementation begins.

License

MIT License - see LICENSE file for details.

Author

Hoseyn Amiri ([e-mail])

Changelog

0.1.0

• Initial placeholder release
• Basic memory storage and recall functionality
• PyPI package structure