python-chgnet

CLEAN
maintainer harrietobrien 0 votes scanned 2026-10-30 00:29:30.650963
View on AUR

Triggered rules

Clean AI review downgraded a static finding llm_review

The static rules flagged this LOW, but an AI model (qwen/qwen3-235b-a22b-2507) reviewed the full PKGBUILD and judged it CLEAN (confidence 95%): The package builds from a source tarball hosted on the project's official GitHub repository, which is a trusted and expected source; all dependencies are standard Python packages, and there is no execution of remote code or use of unverifiable binaries.

1 higher static finding superseded - not the current verdict (shown for transparency)
Low Few votes, recently uploaded zero_votes_recent

Uploaded within the last 14 days with 2 or fewer community votes — little peer review so far.

PKGBUILD

1# Maintainer: Harriet O'Brien <harrietobrien@protonmail.com>
2
3pkgname=python-chgnet
4_pkgname=chgnet
5pkgver=0.4.2
6pkgrel=1
7pkgdesc="Pretrained Universal Neural Network Potential for Charge-informed Atomistic Modeling"
8arch=('x86_64')
9url="https://github.com/CederGroupHub/chgnet"
10license=('BSD-3-Clause')
11depends=(
12 'python>=3.10'
13 'python-pytorch'
14 'python-numpy'
15 'python-pymatgen-core'
16 'python-ase'
17 'python-nvidia-ml-py'
18 'cython'
19 'python-typing_extensions'
20)
21makedepends=(
22 'python-build'
23 'python-installer'
24 'python-setuptools'
25 'python-wheel'
26 'cython'
27)
28source=("${_pkgname}-${pkgver}.tar.gz::https://github.com/CederGroupHub/chgnet/archive/refs/tags/v${pkgver}.tar.gz")
29sha256sums=('753f32c1e30044be85e9c7a126d30f231520c5fdc7a558bf48c369392c247681')
30
31build() {
32 cd "${_pkgname}-${pkgver}"
33 /usr/bin/python -m build --wheel --no-isolation
34}
35
36package() {
37 cd "${_pkgname}-${pkgver}"
38 /usr/bin/python -m installer --destdir="${pkgdir}" dist/*.whl
39 install -Dm644 LICENSE -t "${pkgdir}/usr/share/licenses/${pkgname}/"
40}
41

Scan history

Scanned at (UTC)SeverityRules
2026-10-30 00:29:30 Clean 2
2026-10-29 23:54:04 Low 1

Report a package

Reports go to the AURWatch maintainer (one person) and are read by hand. No login required.

0 / 4000
Your suggestion