Submit, review and improve your work

Use the dedicated submission repository, not a pull request to the curriculum:

https://github.com/labalicious-learning/lab-submissions

Start safely

Session 00 readiness is a private instructor demonstration. For sessions 01–12, first read the submission repository's public-safety policy and step-by-step guide. A public fork, commit or draft PR is already public. Prepare privately, inspect every file and screenshot, and run the preflight before first upload. Automated scans cannot guarantee privacy or erase public history.

Students retain rights in their original work. The course grants permission to use/adapt its materials for coursework; it does not claim ownership of students' contributions. Publication, optional AI-provider review and marketing reuse are separate decisions. Ask the instructor privately for an equivalent local/private review route if publication is inappropriate. Grades, identities, accommodations and private coaching are never posted publicly.

One packet per lab

Your instructor supplies a cohort code and alias privately. In your fork of the submission repo, use cohorts/COHORT/ALIAS/session-NN/ with:

submission.json: session, course commit, alias, explicit public-safety declaration and separate AI consent.
submission.md: outcome, expected/observed behavior, evidence, environment, verification, limits and reflection.
ai-use.md: AI or mock/paired assistance, what you verified, corrections, sources and collaborator aliases.
evidence/: small synthetic text/CSV/JSON or PNG/JPEG evidence.
Optional changes.patch: code changes against the exact course commit. No student code is run automatically.

Reuse this course's evidence packet, charter, task brief and evaluation card inside the packet. The current ignored submissions/ folder remains local scratch space; it is not the public destination.

What to submit by session

Session	Main artifacts
00	Private readiness demonstration; no public account/security checklist
01	Claim classifications, evidence packet, conclusion limits
02	Task brief, verified repository map, test/check result, unresolved question
03	Preserved synthetic raw data, working CSV, sources, decision memo, automation boundary
04	Risk-ranked test charter and first three tests with rationale
05	Contact lookup/allocation/send-permission journey, candidate/role comparison, mobile/API evidence, regression proposal
06	Four-lock read/PATCH-permission/send table, positive control, simulation limits, regression proposal
07	Patch, actual red/green results or explicitly unexecuted assertion design, coverage limits
08	Annotated screenshots, OS/browser/viewport/zoom, keyboard evidence, user impact
09	Implemented/simulated/absent map, healthy/failure/recovery event counts and decision flags, downstream limits
10	Round-by-round recommendations and factual incident update
11	Evaluation card, hypothetical role recommendations, added case; mock unavailable metrics marked not measured
12	Published capstone card, charter, brief, evidence, automation proposal/design, verdict, reflections and changed-condition defense

Group packets include each alias's contribution and reflection. A polished team artifact does not replace individual understanding. Public answers are available: cite sources and collaboration, and be ready to explain one decision and handle a small changed scenario.

Review flow

Draft → public-safety/completeness check → peer reproduction → optional instructor-requested AI coaching → instructor acceptance or revision.

The deterministic bot checks file limits, structure, common credential patterns and relative links without executing artifacts. Green means ready for human inspection, not correct or safe beyond doubt. AI review requires provider setup, separate student consent and instructor approval of the exact commit; otherwise peer and human review remain fully available. AI offers cited coaching, never grades or merges. It cannot claim to have run tests or inspected images.

Revise on the same PR branch. Each commit gets its own review; acceptance of an old commit is not acceptance of a new one. Instructors record scores privately and merge only the reviewed public-safe packet. Do not post numeric grades or learner rankings.

Submission queue: https://github.com/labalicious-learning/lab-submissions/pulls

Instructor workflow: https://github.com/labalicious-learning/lab-submissions/blob/main/INSTRUCTOR.md