Metadata-Version: 2.4
Name: agenticmemory-sdk
Version: 1.0.0
Summary: Official Python SDK for Agentic Memory — 3-tier persistent memory for AI agents. Short-term recall, semantic search, and auto-summarised knowledge in one API.
Author-email: Joe Wee <joe@tyga.cloud>
License: SEE LICENSE IN LICENSE
Project-URL: Homepage, https://agenticmemory.ai
Project-URL: Documentation, https://agenticmemory.ai/docs
Project-URL: Quickstart, https://agenticmemory.ai/quickstart/python
Project-URL: Pricing, https://agenticmemory.ai/pricing
Keywords: ai,agent,memory,llm,langchain,crewai,autogen,semantic-search,vector,3-tier,mcp
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: requests>=2.28
Dynamic: license-file

# agenticmemoryai

Official Python SDK for [Agentic Memory](https://agenticmemory.ai). Give your AI agents persistent memory in one line.

## Install

```bash
pip install agenticmemoryai
```

## Quick Start

```python
from agenticmemoryai import AgenticMemory

memory = AgenticMemory(api_key="amk_your_key")

# Store a message
memory.store("your-space-id", role="user", content="Hello!")

# Recall recent messages
messages = memory.recall("your-space-id", limit=20)

# Search by meaning (Pro plan)
results = memory.search("your-space-id", "pricing discussion")
```

## Features

- **Short-term memory** — instant recall, sub-millisecond reads
- **Medium-term memory** — search past conversations by meaning
- **Long-term memory** — auto-summarised knowledge across months

## API

- `store(space_id, role, content)` — append a message
- `recall(space_id, limit)` — get recent messages
- `search(space_id, query, scope, tags)` — semantic search
- `set_context(space_id, key, value)` — set key-value context
- `get_context(space_id, key)` — get context value
- `add_entry(space_id, type, title, content, tags)` — append entry
- `bootstrap(space_id, ...)` — load full context in one call
- `set_shared(space_id, key, value)` — shared world memory
- `create_space(name, slug)` — create a memory space

## Get an API key

Sign up free at [agenticmemory.ai](https://agenticmemory.ai/auth/signup). 1,000 messages/month on the free tier.

## Links

- [Quick Start Guide](https://agenticmemory.ai/quickstart/python)
- [API Docs](https://agenticmemory.ai/docs)
- [Pricing](https://agenticmemory.ai/pricing)

## Acknowledgements

The structured memory architecture was influenced by work from the [h-network](https://github.com/h-network) community by Halil Baysal.

## License

Proprietary — Tyga.Cloud Ltd. See LICENSE file.
