Metadata-Version: 2.4
Name: actualization-harbor
Version: 0.1.0
Summary: Agent-agnostic training harbor — detect model, adapt flow state, any agent can dock
Author-email: Oracle1 <oracle1@cocapn.ai>
License-Expression: MIT
Project-URL: Homepage, https://github.com/SuperInstance/actualization-harbor
Keywords: actualization,harbor,agent-training,cocapn,adaptive
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: Programming Language :: Python :: 3
Requires-Python: >=3.10
Description-Content-Type: text/markdown

# actualization-harbor

Agent-agnostic training harbor. The harbor doesn't care what kind of ship you are — it adjusts the channel depth to fit your hull.

Any agent can dock, get assessed, receive adapted training flow, and depart more capable. Detects model type, context window, capabilities, and preferred tempo. Generates adaptations automatically.

## Usage

```python
from actualization_harbor import ActualizationHarbor

harbor = ActualizationHarbor()
harbor.register("jetson-agent", model_type="llm-7b", context_window=4096,
                capabilities=["edge", "cuda"])

harbor.enter("jetson-agent")
adaptations = harbor.assess("jetson-agent")
harbor.train("jetson-agent")
harbor.integrate("jetson-agent")
harbor.depart("jetson-agent")
```

Zero deps. `pip install actualization-harbor`
