Test Plan — Personal Training OS (Phase 5: Verification & Validation)

Artifact: V&V test plan · Owner: Delivery Manager (Primary User) · Version: v1.1 · Date: 2026-07-05

Changelog v1.1 (2026-07-05): reworded TC-08, TC-11, TC-19 for testability after defects D-01/D-02/D-03 surfaced ambiguity during run 2026-07-04-run-1 (case-quality fixes, not scope changes). Traces to: spec/requirements/ptos.sdoc (by UID) · Gate: exits at G5 (tests pass + UAT accepted)

⚠️ This is the master baseline — read-only. Do not record run results in the ⬜ cells below. Each execution creates a dated run record under runs/ (copy the template). The ⬜ here mean "case exists," not "not yet run." See runs/README.md.

Verification = "did we build it right?" (against the spec). Validation = "did we build the right thing?" (against user needs). This plan formalizes the checks that were deliberately deferred from Build into runnable test cases. Claude authored this plan; Primary User executes it (with Secondary User for multi-person UAT) — keeping the human-in-the-loop V&V discipline intact.

1. Scope

The live Notion Personal Training OS (page b3bcfff6…, under Personal Dashboard) built in Phase 4. In scope: MVP (Must/Should/Could) behaviour. Out of scope: deferred integrations (Garmin/MFP/Strava/ Calendar sync) and RAG — those are gated later phases (Won't-this-release).

2. Approach & methods

Per requirement verification method: T = Test (do X, assert Y) · I = Inspection (look/confirm) · D = Demonstration (show it works) · A = Analysis. Environment = the live workspace on desktop and the Notion mobile app (for NFR-USE).

3. Entry / exit criteria

• Entry: Phase 4 build complete (G4 passed); this plan reviewed.
• Exit (G5): every priority test case Pass; UAT accepted by Primary User (+ Secondary User for multi-person); any defects triaged (fixed or accepted with rationale).

4. Test cases

Result: ⬜ not run · ✅ Pass · ❌ Fail. Fill Result + Notes as you run each.

Automated run 2026-07-05 — ✅ 10/10 PASS — TC-01, 02, 03, 05, 07, 10, 12, 13, 15, 20 verified via pytest against the live Notion API (verification/tests/; report in report.html). The remaining cases below are manual (inspection / demonstration / mobile / backup) + the UAT section.

ID	Traces	Title	M	Steps	Expected	Result
TC-01	HOK-004/006/008	Hockey NOT attended doesn't count	T	Open Team A game (missed) (Attendance = Did Not Attend)	Counts Toward Training (auto) = unchecked/false	⬜
TC-02	HOK-007	Hockey attended counts	T	Open Team A game (attended)	Counts Toward Training (auto) = checked/true	⬜
TC-03	HOK-004 (else branch)	Completed run counts	T	Open 11-mile long run (Run, Completed)	Counts = true	⬜
TC-04	HOK-004 (else branch)	Planned session doesn't count	T	Create a Run with Status = Planned	Counts = false; then set Status = Completed → true	⬜
TC-05	SHO-004	Shoe mileage rollup	T	Open an example running shoe	Total Miles = 16 (11 + 5 from the two linked runs)	⬜
TC-06	SHO-004	Rollup updates on new run	T	Add a 3-mi run linked to an example running shoe	Total Miles → 19; then delete/unlink → back to 16	⬜
TC-07	PEO-005, TS-005	Secondary User is first-class	T	Create a Training Session; set Person = Secondary User	Secondary User selectable; session saves & shows under her	⬜
TC-08	REL-002	Person filter isolates data	T	In Training Sessions and Metrics, filter Person = Primary User, then = Secondary User	Each filter shows only that person's rows (no cross-leak). Scope: these two person-DBs, not an exhaustive sweep of all DBs	⬜
TC-09	AUTO-004	Manual source default	I	Inspect a hand-created session's Source	Defaults to / set to Manual	⬜
TC-10	EQP-004/005	Equipment availability	I	Open Equipment DB	Kettlebells/Net/Barbell/Medicine ball = Not available; Stickhandling ball = Pending; rest Available	⬜
TC-11	EQP-006	Seeded templates link only Available gear	I	Open the 3 seeded Workout Templates' Equipment relation	The seeded links reference only Available equipment (no barbell/kettlebell/net). Note: a Notion relation can't enforce availability — this verifies the seed data, not a schema constraint	⬜
TC-12	RACE-003/004	Race seeds correct	I	Open Races	SF 1st Half = 2026-07-26, Half Marathon, Goal 2:00:00; Example 5K = 2026-07-25	⬜
TC-13	HIST-001/002	Historical runs correct	I	Open the two seeded runs	11-mi = 2026-06-28 / Long; 5-mi hill repeats = 2026-07-01 / Hills	⬜
TC-14	REL-001, TS-008	Session relationships wired	I	Open 11-mile long run	Related to Person, Running Shoe, Running Route, Training Block ("Example Half Marathon Build")	⬜
TC-15	GOL-003	Goal linked to Race	I	Open goal Sub-2:00 at Example Half Marathon	Race relation = Example Half Marathon; Person = Primary User	⬜
TC-16	HOK-002	Source-of-truth game link	D	On a hockey session, add a Team A Game relation → pick a real game	Links to the Team A Games DB without altering that DB	⬜
TC-17	NAV-002/003	Placement	I	Locate the OS page	Child of Personal Dashboard, link below Inbox	⬜
TC-18	NFR-USE-001	Mobile capture	D	Log a session/check-in on the Notion mobile app	Low-effort capture works on mobile	⬜
TC-19	NFR-BKP-001	Backup/recovery	D	Export the OS page: ••• → Export → format Markdown & CSV, Include subpages = Yes, Create folders for subpages = Yes	Export .zip downloads and contains a .csv per database with its rows (spot-check Training Sessions + Running Shoes). Optional deeper check: re-import the zip into a scratch page and confirm rows land = "restorable"	⬜
TC-20	SC5	Training materials present	I	Open Workout Templates / Hockey Drills	≥1 running + 1 strength + 1 hockey plan; drills present	⬜

5. UAT — validation against user stories

Primary User (and Secondary User) confirm each delivers the intended value (accept / reject + comment).

UAT	Story	Accept?
UAT-1	US1 — every run/race/hockey game in one place	⬜
UAT-2	US2 — capture is low-effort	⬜
UAT-3	US3/SC6 — shoe mileage tracked automatically	⬜
UAT-4	US4 — hockey counts only when attended	⬜
UAT-5	US5 — Secondary User's training tracked (Secondary User confirms)	⬜
UAT-6	US7/SC5 — training plans/drills usable	⬜
UAT-7	US8 — data feels private + is backed up	⬜

6. Defect log

ID	Test	Severity	Description	Status
—			(log failures here)

7. Sign-off — Gate G5

Role	Name	Decision	Date
Delivery Manager / PO	Primary User	✅ APPROVED	2026-07-05

Basis: V&V cycle complete — 20/20 test cases Pass, 7/7 UAT Accept, 0 open defects (see runs/2026-07-04-run-1.md + runs/2026-07-05-summary.md). Gate G5 passed; MVP accepted for release.

Signature: this approval is authenticated by the SSH-signed commit that records it and the signed vv-accepted tag marking the accepted baseline. Verify with git tag -v vv-accepted.

On sign-off: git tag vv-accepted, then proceed to Phase 6 (Release).