Metadata-Version: 2.5
Name: vantadb-letta
Version: 0.5.0
Summary: Letta adapter for VantaDB — embedded vector memory for AI agents
Project-URL: Homepage, https://github.com/ness-e/Vantadb
Project-URL: Repository, https://github.com/ness-e/Vantadb
Project-URL: Documentation, https://github.com/ness-e/Vantadb/tree/main/integrations/letta
Author: VantaDB Team
License: Apache-2.0
Keywords: ai-memory,letta,rag,vantadb,vector-store
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Database
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.11
Requires-Dist: letta-client<2,>=1.0.0
Requires-Dist: vantadb-py<0.9.0,>=0.6.1
Description-Content-Type: text/markdown

# VantaDB × Letta

Letta storage adapter for [VantaDB](https://github.com/ness-e/Vantadb).

## Install

> **Not on PyPI yet.** The `vantadb-letta` package declares Alpha
> status and will go live with the first `adapters-v*` tag release
> (no local `dist/` built yet). Until then, install from source.

```bash
# Today, from a repo checkout
cd integrations/letta && pip install .
```

### Install from PyPI (after first release)

```bash
pip install vantadb-letta
```

## Quickstart

```python
from vantadb_letta import VantaDBVectorStore

store = VantaDBVectorStore(
    db_path="./my_data",
    namespace="agent_memory",
)

store.insert("VantaDB is a vector database written in Rust.", source="docs")
results = store.search("vector database")
for r in results:
    print(r["text"])
```

## API

- `insert(text, source=None, metadata=None)` — store a memory record
- `search(query, k=4)` — search stored memories
- `delete(key)` — delete by key
- `list(limit=100)` — list all records

## Status: experimental

Letta is a stateful platform with its own memory layer and no public
vector-store contract; this adapter is a community convenience, not an
officially supported integration. The API may change without notice and it
is not exercised by Letta's own test suite. Prefer Letta's native memory
unless you specifically need to back it with an embedded local store.

## Why VantaDB?

- **Embedded & local-first:** the storage engine is a Rust library embedded
  in your process — no server to deploy, no network hop; data lives in your
  filesystem.
- **Persistent hybrid search:** vectors + BM25 text search out of the box,
  where Letta's own memory is designed for its server-side platform.
- **Zero-setup alternative to hosted stacks:** unlike Zep (requires a server)
  or Cognee (spins up its own knowledge-graph runtime), VantaDB is a plain
  library you import.

## Development

```bash
pip install -e .
```
