Metadata-Version: 2.4
Name: cortex-ai-mem
Version: 1.0.0
Summary: Persistent memory for AI agents — Python SDK
Project-URL: Homepage, https://cortex-kappa-ten.vercel.app
Project-URL: Documentation, https://cortex-kappa-ten.vercel.app/docs
License-Expression: MIT
Keywords: agent,ai,cortex,mcp,memory
Requires-Python: >=3.9
Requires-Dist: requests>=2.28
Description-Content-Type: text/markdown

# cortex-memory

Persistent memory for AI agents — Python SDK.

## Install

```bash
pip install cortex-ai-mem
```

## Usage

```python
from cortex_memory import CortexMemory

memory = CortexMemory(api_key="ctx_your_key_here")

# Save
memory.save("User prefers dark mode and Python", tags=["preference"])

# Search (semantic)
results = memory.search("theme preferences")

# Update
memory.update(results[0]["id"], "User prefers dark mode, Python, and VS Code")

# Delete
memory.delete(results[0]["id"])

# List all
all_memories = memory.list()
```

## Get an API key

1. Go to [cortex-kappa-ten.vercel.app](https://cortex-kappa-ten.vercel.app)
2. Sign in with GitHub or Google
3. Navigate to Dashboard → Keys
4. Create a key (starts with `ctx_`)

## Features

- **Semantic search** — find memories by meaning, not keywords
- **User isolation** — each API key has its own private memory space
- **Tags & filtering** — organize by agent, session, or custom tags
- **Works with any framework** — Strands, LangChain, CrewAI, or plain Python
