Metadata-Version: 2.4
Name: pryvra-pipecat
Version: 0.1.1
Summary: Pryvra privacy boundary processor for Pipecat pipelines.
Author: Pryvra
License: Apache-2.0
Project-URL: Homepage, https://github.com/sahil-rajput/pryvra
Project-URL: Repository, https://github.com/sahil-rajput/pryvra
Project-URL: Issues, https://github.com/sahil-rajput/pryvra/issues
Keywords: pipecat,pii,privacy,llm,voice-agents
Classifier: Development Status :: 3 - Alpha
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Requires-Python: >=3.10
Description-Content-Type: text/markdown
Requires-Dist: pipecat-ai>=0.0.1

# Pryvra for Pipecat

Install the processor in your Pipecat application:

```bash
pip install pryvra-pipecat
```

It uses Pipecat's normal frame-processor contract and expects a vault object
with `tokenize(text).text` and `restore(text)` methods. The vault can be a
Python client that calls the trusted service holding your Pryvra AES key.

```python
from pryvra_pipecat import PrivacyFrameProcessor

pipeline = Pipeline([
    transport.input(), stt,
    PrivacyFrameProcessor(vault, mode="input"),   # before context aggregator / LLM
    context_aggregator.user(), llm,
    PrivacyFrameProcessor(vault, mode="output"),  # before TTS
    tts, transport.output(), context_aggregator.assistant(),
])
```

Keep the two processors within the same trusted service boundary as the vault.

For local development before the first PyPI release:

```bash
pip install ./integrations/pipecat
```
