menonpg

menonpg / soul.py

Public

Persistent identity and memory for any LLM agent — markdown-native, provider-agnostic

33
1
100% credibility
Found Mar 02, 2026 at 20 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

soul.py is a lightweight Python tool that adds persistent memory to AI assistants using everyday text files, so they remember past conversations without any complex setup.

How It Works

1
🧠 Discover soul.py

You hear about soul.py, a simple way to give your AI a lasting memory that sticks around even after chats end.

2
💻 Set up your assistant

You easily add soul.py to your computer and create two plain text files for your AI's personality and memories.

3
✨ Define your AI's soul

You write a short description in the soul file to shape your AI's personality and how it behaves.

4
🔗 Connect an AI thinker

You link soul.py to a smart AI service so your assistant can understand and respond to you.

5
💬 Start chatting

You ask your AI questions or share facts about yourself, and it saves them to remember later.

6
😲 Magic memory test

In a brand new chat, you ask what it knows about you, and it recalls everything perfectly.

🌟 Forever remembering companion

Your AI now grows with you, remembering every conversation across sessions like a true friend.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 20 to 33 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is soul.py?

soul.py adds persistent identity and memory to LLM agents using markdown files—no databases or servers needed. Your AI remembers facts, names, and context across sessions, working with Anthropic, OpenAI, or any provider via simple Python APIs. Run `soul init` for setup, then ask questions that stick, like a github persistent identifier for stateless chats.

Why is it gaining traction?

It ditches framework bloat like LangChain for a lightweight primitive: human-readable markdown you can git-track, zero lock-in, and auto-routing between fast semantic search and deep synthesis. Demos show memory persisting in new processes, plus timings for RAG/RLM paths. Developers hook it into custom agents without overhead.

Who should use this?

AI builders prototyping persistent agents, like research lab tools or chatbots needing user history. Indie devs avoiding github persistent storage complexity for quick LLM memory. Teams wanting versioned, markdown-native recall over msix persistent identity hacks.

Verdict

Grab it for simple persistence (19 stars, solid README/demos), but 1.0% credibility flags early maturity—test in prototypes, not prod. Promising if you need github persistent windows without infra.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.