Metadata-Version: 2.4
Name: gwatch
Version: 0.0.20
Summary: G-Watch is a toolbox for GPU profiling and program analysis.
Home-page: https://github.com/G-Watch/G-Watch
Author: Zhuobin Huang
Author-email: zhuobin@u.nus.edu
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: prettytable
Provides-Extra: test
Requires-Dist: pytest; extra == "test"
Dynamic: author
Dynamic: author-email
Dynamic: classifier
Dynamic: description
Dynamic: description-content-type
Dynamic: home-page
Dynamic: license-file
Dynamic: provides-extra
Dynamic: requires-dist
Dynamic: requires-python
Dynamic: summary

# G-Watch

![nvidia](https://img.shields.io/badge/NVIDIA-Supported-green.svg?logo=nvidia)
![amd](https://img.shields.io/badge/AMD-Supported-red.svg?logo=amd)

**G-Watch** is a toolbox for agentic GPU kernel optimization.
It features rich **Profiling** capabilities on both NVIDIA and AMD GPUs.
Additionally,
G-Watch offers **Program Analysis** tools for inspecting compiler-generated GPU binaries,
facilitating secondary development tasks such as register analysis and binary instrumentation.

## Installation

```bash
# install gwatch
pip install gwatch

# install gwatch skills for codex/claude/gemini-cli
gwatch install_skill
```

## Usage

In your Code Agent, use G-Watch skills, and run

```bash
# the auto optimization skill
/gwatch_cuda_auto_optimization

# brief description of the workload
workload is under xxx/xxx, i want to optimize the performance of xx operator/kernel
```

## Links

- [GitHub](https://github.com/mars-compute-ai/G-Watch)
- [Documentation](https://github.com/mars-compute-ai/G-Watch)
