Metadata-Version: 2.4
Name: newb
Version: 0.4.0
Summary: Test your package through the eyes of a newbie agent — a fresh AI agent reads only your docs/skills and tries to use your package.
Author-email: Yusuke Watanabe <ywata1989@gmail.com>
License-Expression: AGPL-3.0-only
Project-URL: Homepage, https://github.com/ywatanabe1989/newb
Project-URL: Repository, https://github.com/ywatanabe1989/newb
Keywords: agentic-test,newbie-test,skill-verify,documentation-testing,ai-agent,llm
Classifier: Development Status :: 3 - Alpha
Classifier: Programming Language :: Python :: 3
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: click>=8.0
Provides-Extra: yaml
Requires-Dist: PyYAML>=6.0; extra == "yaml"
Dynamic: license-file

# newb

<p align="center"><img src="./assets/newb-logo.png" width="220" alt="newb mascot"/></p>

A fresh AI agent reads only your `_skills/` (or equivalent docs) and tries
to use your package. If it succeeds, your docs work.

## Install

```bash
pip install newb
```

## Use

```bash
newb ./src/mypkg/_skills/mypkg
newb ./_skills --format markdown >> README.md
```

Spins up a clean docker container with **only your skills** mounted, then
asks a fresh Claude agent four canonical questions (what for / problems /
quick start / when not to use), plus optional red tests from
`_red_tests.yaml`.

Output: JSON (for CI) or markdown (for README injection).

## Library

```python
import newb
report = newb("./src/mypkg/_skills/mypkg")
print(newb.render_markdown(report))
```

## Requirements

- Docker on PATH
- `ANTHROPIC_API_KEY` in env
- Python 3.10+

## License

AGPL-3.0-only.
