Metadata-Version: 2.4
Name: usetrace
Version: 0.1.0
Summary: Lightweight tracing SDK for LLM applications
Project-URL: Homepage, https://github.com/sanyamsharma/trace
Project-URL: Repository, https://github.com/sanyamsharma/trace
Project-URL: Issues, https://github.com/sanyamsharma/trace/issues
Author: Sanyam Sharma
License-Expression: Apache-2.0
License-File: LICENSE
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Software Development :: Libraries
Classifier: Topic :: Software Development :: Testing
Classifier: Typing :: Typed
Requires-Python: >=3.11
Requires-Dist: httpx>=0.27.0
Requires-Dist: pydantic>=2.0.0
Provides-Extra: dev
Requires-Dist: pytest-asyncio>=0.24.0; extra == 'dev'
Requires-Dist: pytest>=8.0.0; extra == 'dev'
Requires-Dist: ruff>=0.8.0; extra == 'dev'
Description-Content-Type: text/markdown

# usetrace

Lightweight tracing SDK for LLM applications. Add `@tracer.observe()` to your functions and get visual attribution maps that answer **"Why did the LLM say that?"**

## Install

```bash
pip install usetrace
```

## Quick Start

```python
from usetrace import Trace

tracer = Trace(api_key="your-key", base_url="https://api.use-trace.com")

@tracer.observe(span_type="llm", model="gpt-4o")
def my_llm_function(prompt: str) -> str:
    return openai.chat.completions.create(...)
```

Every traced call is captured — inputs, outputs, latency, token usage — and sent to the Trace dashboard where you can inspect full execution trees and per-token attribution maps.

## Dashboard

Sign up and explore your traces at [use-trace.com](https://use-trace.com).

## License

Apache-2.0
