Skip to content

CONTRIBUTING.md

Latest commit

 

History

History
80 lines (63 loc) · 3.42 KB

CONTRIBUTING.md

File metadata and controls

80 lines (63 loc) · 3.42 KB
Raw

Contributing to Canvas Pilot

Thanks for your interest. Canvas Pilot is a Claude Code project: the behavior lives in Markdown skill files under .claude/skills/, and the Python in src/ is API-primitive glue (Canvas client, PDF tooling, verifiers). Most contributions are either (a) framework Python / hook fixes or (b) skill-prose improvements.

By participating in this project you agree to abide by its Code of Conduct.

Ground rules — read these or your PR will be rejected on sight

This repo ships under a strict public/private split.

  • Never commit personal or school-identifying data. Course IDs, file IDs, instructor names, school names, emails, cookies, JWTs, real drafts — none of it. All of that lives in gitignored paths (_private/, SECRETS.md, courses.yaml, .env, .cookies/, sources/, runs/). A PreToolUse hook (.claude/hooks/check-public-leak.py) blocks commits that add known identity literals, but the hook is a backstop, not a license to be careless.
  • git add always names explicit paths. Never git add -A, git add ., or git add -f. Verify with git diff --cached before committing.
  • Never edit .gitignore to remove a safety entry, even "just for testing".
  • No school-specific or instructor-specific solving logic in tracked files. The tracked skills are generic frameworks; per-school behavior belongs in a local _private/canvas-<name>-app.md overlay that never leaves your machine.

Dev setup

git clone https://github.com/X-isdoingreat/Canvas_pilot.git
cd Canvas_pilot
python -m venv .venv
. .venv/bin/activate          # Windows: .venv\Scripts\activate
pip install -r requirements.txt
python -m playwright install chromium    # one-time, ~150 MB

Open the folder in Claude Code and say scan canvas to exercise the entry flow. On a fresh checkout this dispatches canvas-setup (asks for your Canvas URL, logs you in once).

Project shape

  • .claude/skills/canvas-*/SKILL.md — the actual behavior. Prose instructions to Claude, organized as pipelines (stages), not monoliths.
  • .claude/hooks/*.py — deterministic guardrails (schema validation, approval gates, leak checks). These run inside Claude Code sessions.
  • src/*.py — API primitives and verifiers. Pure-ish Python, unit-testable.
  • scripts/*.py — cron-able helpers (reminders, autonomous submit) for Windows Task Scheduler / /schedule.
  • tests/ — these are Codex-sidecar runners (run_*.py), not pytest suites. Run a relevant one directly, e.g. python tests/codex_runtime/run_all.py.

Tests

Run the runner for the area you touched:

python tests/codex_runtime/run_all.py        # scan / execute / router / onboarding
python tests/codex_hooks/run_hook_tests.py   # hooks

Pull requests

  • Keep PRs scoped to one concern.
  • Describe what changes for the student ("the scanner now…"), not just the diff.
  • If you touched a hook or the public/private boundary, say so explicitly and confirm check-public-leak.py still blocks the bad cases.
  • By contributing, you agree your contribution is licensed under AGPL-3.0-or-later, the project license.

Questions / disclosure

  • General questions: open an issue, or reach the maintainer on X @X_isdoingreat.
  • Security or accidental-leak reports: see SECURITY.md. Do not open a public issue for those.