| name | ci |
| description | missense-kinase-toolkit-specific CI/CD layout; extends my central `ci` skill. Use when editing .github/workflows, adding a sub-package, changing the test env, or debugging CI failures. |
missense-kinase-toolkit CI/CD
Baseline — fetch first
Apply my canonical ci conventions (path-filtered per-package workflows,
micromamba env, OS/Python matrix, Codecov flags, pre-commit hook set) before the
repo-specific notes below. WebFetch and follow:
https://raw.githubusercontent.com/jessicaw9910/skills/main/.claude/skills/ci/SKILL.md
If the fetch fails (no network / non-200), tell me the central ci skill
could not be retrieved and confirm how to proceed — do not silently skip the
baseline.
Repo-specific additions
Workflows (.github/workflows/)
One workflow per setuptools sub-package, path-filtered (plus a weekly
cron: "0 0 * * 0"):
schema-ci.yaml — paths missense_kinase_toolkit/schema/**, coverage flag
schema, installs only the schema package.
databases-ci.yaml — paths missense_kinase_toolkit/databases/**, coverage
flag databases, installs schema then databases (--no-deps, because
databases depends on schema). Runs pytest with -n 2 --dist loadfile.
Both: matrix os: [macOS, ubuntu, windows] × python: ["3.10", "3.11"],
mamba-org/setup-micromamba@v1 with
environment-file: devtools/conda-envs/test_env.yaml, then Codecov upload with
the per-package flags. There is no workflow for ml/, experiments/, or
app/.
Adding / changing
- New runtime/test dependency → add it to
devtools/conda-envs/test_env.yaml
(CI installs the package with --no-deps, so deps must be in the env file).
Pre-commit
.pre-commit-config.yaml is scoped to ^missense_kinase_toolkit (excluding
KinaseInfo/): black, isort (profile black), flake8 (max-line-length 88,
ignore E203/E501), pyupgrade --py39-plus, plus whitespace/yaml hooks.