Metadata-Version: 2.4
Name: klaket
Version: 0.7.1
Summary: Python client + CLI for Klaket — turn any video into LLM-ready data
Author: Klaket
License-Expression: MIT
Project-URL: Homepage, https://github.com/huseyinstif/klaket
Project-URL: Repository, https://github.com/huseyinstif/klaket
Project-URL: Issues, https://github.com/huseyinstif/klaket/issues
Keywords: video,llm,transcription,whisper,diarization,subtitles,mcp
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Multimedia :: Video
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Dynamic: license-file

# klaket (Python SDK)

Turn any video into LLM-ready data. Zero dependencies.

```bash
pip install klaket
```

```python
from klaket import Klaket

k = Klaket()  # self-hosted; Klaket("https://api.klaket.dev", api_key="klk_...") for cloud

result = k.process(
    "https://youtube.com/watch?v=...",
    model="medium",                 # optional quality bump
    prompt="Klaket, Docker, RAG",   # proper-noun hints
    num_speakers=2,                 # diarization hint
)

for seg in result["transcript"]:
    print(f'[{seg["start"]:7.2f}] {seg.get("speaker", "")}: {seg["text"]}')

hits = k.search(result["id"], "docker compose command")
```

MIT licensed. Part of [Klaket](https://github.com/klaket/klaket).
