Metadata-Version: 2.4
Name: swiss-ai-hub-agent
Version: 0.291.5
Summary: Swiss AI Hub Agent SDK: build transparent, workflow-based, event-driven AI agents.
Author: Joel Barmettler, Marius Högger, Michèle Fundneider, Thomas Mannhart, Noah Hermann
Author-email: Joel Barmettler <joel.barmettler@bbv.ch>, Marius Högger <marius.hoegger@bbv.ch>, Michèle Fundneider <michele.fundneider@bbv.ch>, Thomas Mannhart <thomas.mannhart@bbv.ch>, Noah Hermann <noah.hermann@bbv.ch>
License-Expression: Apache-2.0
License-File: LICENSE
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Requires-Dist: swiss-ai-hub-core==0.291.5
Requires-Dist: fastmcp>=3.0.0
Requires-Dist: llama-index-llms-openai>=0.6.12
Requires-Dist: llama-index-llms-azure-openai>=0.4.2
Requires-Dist: stringcase>=1.2.0
Requires-Python: >=3.13, <3.14
Project-URL: Homepage, https://github.com/bbvch-ai/aihub-core
Project-URL: Repository, https://github.com/bbvch-ai/aihub-core
Project-URL: Documentation, https://bbvch-ai.github.io/aihub-core/
Project-URL: Issues, https://github.com/bbvch-ai/aihub-core/issues
Description-Content-Type: text/markdown

# Swiss AI Hub Agent

Agent SDK for the [Swiss AI Hub](https://github.com/bbvch-ai/aihub-core) platform. Build transparent, workflow-based,
event-driven AI agents that run as independent microservices over NATS/JetStream.

- **Custom workflow engine** — define steps with `@step()`; inputs and outputs are inferred from event type annotations
  and dispatched via NATS, so consecutive steps can execute on different servers.
- **Stateless & decentralized** — all state lives in Redis (`RunContext`/`ThreadContext`) and the JetStream event
  history; agents scale horizontally.
- **Batteries included** — config with Form duality, dependency injection, i18n, memory, LLM streaming, and a catalog of
  pre-built agents (RAG, expert-asking, few-shot, …).

## Installation

```bash
pip install swiss-ai-hub-agent
```

This pulls in [`swiss-ai-hub-core`](https://pypi.org/project/swiss-ai-hub-core/).

## Usage

```python
from swiss_ai_hub.agent import Agent, AgentRunner
```

## Links

- Source & issues: https://github.com/bbvch-ai/aihub-core
- Documentation: https://bbvch-ai.github.io/aihub-core/

## License

Apache-2.0
