Context
There's no CONTRIBUTING.md at the repo root. New contributors
have to read the README, the test docstring, and pyproject.toml
to figure out: how to run the tests, where to log an issue,
the PR conventions (commit message style, branch naming, "no
ISSUES.md in commits"), and how to add a new model id safely.
The README covers most of it implicitly but not for contributors.
What to change
Create CONTRIBUTING.md at the repo root with four short
sections:
- Setup — clone, venv,
pip install -r requirements.txt,
python -m unittest discover -s tests -v.
- Tests — the dual-runner docstring note (pytest + unittest),
the STREAMLIT_RUN=1 env var, the numpy/pandas skip behaviour.
- PR conventions — one commit per fix, brief messages,
ISSUES.md is personal tracking and must NOT be committed.
- Adding a new model — the
curl recipe from
app_helpers.AVAILABLE_MODELS docstring, plus the test
to update.
How to verify
head -5 CONTRIBUTING.md describes setup in one paragraph.
- The doc references the actual commands, not paraphrases.
- A new contributor can clone, install, run tests, and open a
PR using only this file (no Slack, no asking around).
Skill: docs.
Estimated effort: S.
Context
There's no
CONTRIBUTING.mdat the repo root. New contributorshave to read the README, the test docstring, and
pyproject.tomlto figure out: how to run the tests, where to log an issue,
the PR conventions (commit message style, branch naming, "no
ISSUES.md in commits"), and how to add a new model id safely.
The README covers most of it implicitly but not for contributors.
What to change
Create
CONTRIBUTING.mdat the repo root with four shortsections:
pip install -r requirements.txt,python -m unittest discover -s tests -v.the
STREAMLIT_RUN=1env var, the numpy/pandas skip behaviour.ISSUES.mdis personal tracking and must NOT be committed.curlrecipe fromapp_helpers.AVAILABLE_MODELSdocstring, plus the testto update.
How to verify
head -5 CONTRIBUTING.mddescribes setup in one paragraph.PR using only this file (no Slack, no asking around).
Skill: docs.
Estimated effort: S.