Introduction
memory plus provides a local RAG‑backed memory store for your MCP agent, so it can record, retrieve, list, and visualize "memories" (notes, ideas, context) across sessions. Think of it as a lightweight personal knowledge base that your agent consults and updates automatically.
🏆 This repo won First Place in the Infosys Cambridge AI Centre Hackathon!
Features
- Record Memories: Persist user data, ideas, or important context for future runs.
- Retrieve Memories: Keyword‑ or topic‑based search over past entries.
- Recent Memories: Quickly see the last N stored items.
- Update Memories: Update existing memories with new information.
- Visualize Memory Graph: Interactive clusters showing how memories interrelate.
Prerequisites
# Create & activate a virtual environment
python3 -m venv .venv
source .venv/bin/activate
Installation
# Install uv if you haven't already
pip install uv
# Run the memory server using uvx
uvx memory-server
Usage
The server provides several tools:
record: Record new memoriesretrieve: Search for relevant memoriesrecent: Get recent memoriesupdate: Update existing memoriesvisualize: Create memory visualizationset_whether_to_annonimize: Toggle anonymization
Configuration
Add the memory server to any MCP‑capable client by adding this to your JSON config:
{
"servers": {
"memory_server": {
"type": "stdio",
"command": "fastmcp",
"args": ["run", "absolute/path/to/memory.py"]
}
}
}
🧠 One-Line Import
If you have past information—such as emails, chats, or AI conversations—that you'd like to store in the memory server, simply export them as a text file and run the following one-liner:
python load_text_to_memory.py path/to/your/previous_memory.txt
RoadMap
- Memory Update
- Improved prompt engineering for memory recording
- Better Visualization of Memory Graph
- Possible Graph Database Integration
- Neon/Supabase Integration
- Web UI for Memory Management
License
This project is licensed under the MIT License. See LICENSE for details.
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file memory_plus-0.1.1.tar.gz.
File metadata
- Download URL: memory_plus-0.1.1.tar.gz
- Upload date:
- Size: 1.7 MB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.11.5
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
c7bbd24058ec6ee714d5e91021b8d6ae427e707fbada02409b351bf0e130815d
|
|
| MD5 |
f2cb2702569aa7c1fdb29fb79d5a1081
|
|
| BLAKE2b-256 |
c04fa592d6378f227e571db96621410b81c22d9e44c0a328d6a74fbafef78a2f
|
File details
Details for the file memory_plus-0.1.1-py3-none-any.whl.
File metadata
- Download URL: memory_plus-0.1.1-py3-none-any.whl
- Upload date:
- Size: 11.1 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.11.5
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
cdf4ed8e954bef308a6f5f7213361cd379918e729b997da4e9a3836cd7bb43ea
|
|
| MD5 |
4c94ae3c9fd5cff26d34100931f6b514
|
|
| BLAKE2b-256 |
bb5144d3e61bf8032babd38f1b146e2c25ba77cb9345b8a9d34fcce859d5b547
|