Metadata-Version: 2.4
Name: scrambo
Version: 0.1.0a8
Summary: Thin Python SDK for Scrambo's cloud-hosted video editing agents
Author: Scrambo
License-Expression: MIT
Project-URL: Homepage, https://scrambo.dev
Classifier: Development Status :: 3 - Alpha
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Requires-Python: >=3.11
Description-Content-Type: text/markdown
Requires-Dist: httpx<1,>=0.27
Requires-Dist: python-dotenv<2,>=1.0

# Scrambo

Scrambo is a thin Python SDK for building reliable video agents. Users can orchestrate specialist workers, let them inspect and validate the actual timeline, and turn one-off prompts into repeatable production workflows, without building the entire video-agent stack from scratch.

Check out the documentation to start: https://www.scrambo.dev/docs/start-here

Agent-only capabilities are imported from `scrambo.tools` and explicitly
granted for one call. `transcribe` and `detect_events` work with Source Work,
Planner Compile, or Author; `detect_beats` works with Source Work. Paid
`masking` works with Source Work, Generate, Planner Compile, Author, Graphics,
Sound, Titles, and Captions when
explicitly granted as `tools=[masking]`; it has no caller configuration and is
not directly callable. Event results include
request-scoped taxonomy IDs when the
prompt asks for specific event types, or unlabeled grounded discoveries for
open-ended requests. The capability prepares timestamped contact sheets; the
selected specialist provider watches them and derives the events.


## Examples: 

```python
from scrambo import editor, planner, source, timeline
from scrambo.tools.genAI import img2video, voiceover

editor.open(project="listing-tour", input="./listing-stills")
brief = source.generate_agent.create(
    "Create a narrated tour",
    tools=[img2video, voiceover],
    budget_usd=5.0,
    max_calls=4,
)
direction = planner.ask("Explore three narrative directions")
plan = planner.compile(direction, "Ground the strongest direction as a 30-second tour")
editor.start()
timeline.author_agent.edit(plan)
```
