Metadata-Version: 2.4
Name: reca11-memory
Version: 0.1.1
Summary: Memory layer and SDK for LLMs
Author-email: Shashi Verma <shashi.shekhar.s.verma@vanderbilt.edu>
Requires-Python: >=3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: fastapi>=0.116.1
Requires-Dist: openai>=1.98.0
Requires-Dist: pinecone>=7.3.0
Requires-Dist: python-dotenv>=1.1.1
Requires-Dist: slowapi>=0.1.9
Requires-Dist: uvicorn>=0.35.0
Requires-Dist: requests>=2.0.0
Requires-Dist: pymongo>=4.13.2
Requires-Dist: build>=1.3.0
Requires-Dist: twine>=6.1.0
Dynamic: license-file

# Reca11 — Memory Layer for LLMs

**Reca11** is a plug-and-play memory architecture designed to add persistent context to your LLM applications. It intelligently tracks and stores three core components of every conversation:

- **Recent chat history**  
- **Core user facts**  
- **Evolving thematic summaries**

This enables better recall, continuity, and personalization in any app powered by large language models.

**Link to research article:** COMING SOON!

---

## Installation

Install the SDK via pip:

```bash
pip install reca11-memory
```

---

## Generate an API Key

To use the memory service, you’ll need an API key.

Visit: [https://reca11-memory-1.onrender.com](https://reca11-memory-1.onrender.com)  
Click the **"Generate API Key"** button to instantly receive a unique key.

---

## Quickstart

Here’s how to get up and running in just a few lines:

```python
from reca11 import Reca11

rc = Reca11(
    api_key="your-api-key",
    openai_key="your-openai-key",
    project_name="my-project"
)

chat_pair = {
    "assistant": "Hi! What did you do today?",
    "user": "I tried pineapple on pizza. Safe to say I'm not a fan!"
}

memory = rc.recall(chat_pair)

print(memory)
```

---

## Health Check

To verify the backend is online:

```
GET https://reca11-memory.onrender.com/health
```

**Expected response:**

```json
{ "status": "ok" }
```

---

## License

This project is open-sourced under the terms of the MIT License.
