Skip to content

Install ytscrape via pip or uv in Your Python Project

Install ytscrape via pip or uv, verify the setup, review the three runtime dependencies (requests, pycountry, defusedxml), and configure a dev environment.

ytscrape is a pure-Python library published on PyPI under the MIT license. It has a minimal footprint — three small runtime dependencies — and works on every major operating system. This page covers everything you need to get it running in your environment.

Requirements

  • Python 3.10 or newer. ytscrape uses modern type annotations and structural pattern features that require Python 3.10+. It is tested against CPython 3.10, 3.11, 3.12, 3.13, and 3.14.
  • Runtime dependencies — installed automatically by pip or uv:
Package Purpose
requests >=2.28 All HTTP communication with YouTube's InnerTube endpoints.
pycountry >=22.3.5 Validates ISO 639-1 language codes and ISO 3166-1 country codes passed to language= and region=.
defusedxml >=0.7.1 Safe XML parsing for transcript/caption track responses.

No browser, no driver, and no external binaries are required.

Optional: async extra

The asynchronous API (AsyncYouTube, AsyncInnerTubeClient) needs httpx. Install it via the package extra:

pip install "ytscrape[async]"

uv add "ytscrape[async]"
Extra package Purpose
httpx >=0.27 Async HTTP client used by AsyncInnerTubeClient (concurrency + backoff).

The base install stays free of httpx if you only use the synchronous YouTube API. See the Async API guide for usage.

Install the Package

pip install ytscrape

uv add ytscrape

git clone https://github.com/vsmutok/ytscrape
cd ytscrape
pip install .

Note

The package name on PyPI is ytscrape — install it exactly as shown above. After installation a ytscrape console script is also registered, so you can run ytscrape search "python" --max 10 directly from your terminal.

Verify the Installation

Open a Python interpreter or create a small script and confirm the package imports correctly:

import ytscrape

print(ytscrape.__version__)  # e.g. 0.1.4

You can also run the built-in CLI to confirm end-to-end connectivity:

ytscrape search "python" --filter videos --max 3

If the command prints video titles and URLs, ytscrape is installed and communicating with YouTube successfully.

Optional: Development Dependencies

If you want to contribute to ytscrape or run its test suite locally, sync the full development environment using uv:

git clone https://github.com/vsmutok/ytscrape
cd ytscrape
uv sync --dev

The dev dependency group includes:

Package Purpose
pytest >=9.1.1 Test runner — execute with uv run pytest.
ruff >=0.14.5 Linting and code formatting.
bandit >=1.9.4 Static security analysis.
pre-commit >=4.3.0 Git hook runner that executes ruff and bandit before every commit.
twine >=6.2.0 Package upload tooling (maintainers only).

Run the full check suite with:

uv run pytest                            # run the test suite
uv run pre-commit run --all-files        # ruff + format + bandit

Next Steps

With ytscrape installed, follow the Quickstart to write your first search, metadata fetch, and comment collection in minutes. For concurrent scraping with asyncio, install ytscrape[async] and read the Async API guide.