← BBS SG Bank submission · All documents

07 — Submission Checklist

Gate list against the official rubric. A box is only checked when the evidence exists and is linked. Several boxes are intentionally unchecked — that is the honest state.

Legend: [x] done + evidence · [~] partial · [ ] not done / blocked.

A. Eligibility gates (mandatory — missing any one is disqualifying)

	Requirement	Status	Evidence / placeholder
A1	Hosted working prototype URL	`[x]` ✅	Interactive Worker (JavaScript port; COBOL terminal runs locally)
A2	Short demo (≤3 min)	`[x]` ✅	`https://bbs-sg-bank-demo.pages.dev/media/bbs-sg-bank-demo.mp4`; script ready `05-demo-script.md`
A3	Before/after explanation	`[x]` ✅	`02-before-after.md`
A4	Original public post per team member (Medium or LinkedIn)	`[ ]` ⛔	`06-social-posts.md`; URLs `{MEMBER_n_POST_URL}`
A5	Built substantially with Opsera Forge	`[~]` 🟡	project created; Assessment complete (57/100, 10 findings; raw report not exported; `docs/forge/assessment.md` is a team transcript); Intent v1 + PRD-Spec v1 + Architecture v1 + User Stories v1 approved and exported, Testing v1 generated and exported (not approved) to `docs/forge/`; Delivery pending (`04-...`)
A6	Public repo	`[x]` ✅	`https://github.com/vasanthsreeram/bbs-sg-bank-modernization`

B. Rubric readiness

B1 — Working prototype (35%)

	Item	Status	Evidence
B1.1	Legacy batch in repo	`[x]` ✅	`legacy/bank.cob`
B1.2	Modern batch implemented (with malformed-row guard)	`[x]` ✅	`modern/bank.py` L23-25
B1.3	End-to-end run produces correct counts/balances	`[x]` ✅	`03-...` §3 (PROCESSED=3, REJECTED=4, TOTAL=230)
B1.4	Automated tests pass	`[x]` ✅	`scripts/test_bank.py` → Ran 33 tests … OK; `tests/test_modern.py` → Ran 5 tests … OK
B1.5	Acceptance tests AT-1…AT-13 pass	`[x]` ✅	`03-...` §4
B1.6	Parity AT-12 (legacy ↔ modern, byte-identical)	`[x]` ✅	`parity … byte-identical: YES` in all `scripts/evidence/benchmark-*.txt`
B1.7	Demo app runs locally	`[x]` ✅	`site/backend/app.py` (50 site tests), smoke-tested 2026-09-23
B1.8	Terminal runs the real COBOL batch	`[x]` ✅	`legacy-ui/` (20 tests): live 1,200×400 job → 400 processed, 3 rejected, 1.197 s actual COBOL execution (1.234 s overall), modern `TOTAL_CENTS` 5489403400
B1.9	Hosted, clickable prototype (public URL)	`[x]` ✅	https://bbs-sg-bank-demo.s-vasanthrojin.workers.dev/

B2 — Substantial Opsera Forge use (30%)

	Item	Status	Evidence
B2.1	Forge project created	`[x]` ✅	`projects/6de0abcc-…` from `/tmp/bbs-sg-legacy.zip` (COBOL SHA-256 `f4730e6d…afc21d` at upload)
B2.2	Assessment / ForgeScore complete	`[x]` ✅	57/100 (Developing); 10 findings incl. 3 high (`O(m×n)` scans, full-file rewrites, missing specs) — read from the signed-in UI 2026-09-23
B2.3	Assessment artifact exported	`[~]` 🟡	`docs/forge/assessment.md` is a team transcription of the observed values (57/100, 10 findings incl. 3 high); the raw report is not exported (MCP 401; browser download available)
B2.4	Intent artifact exported	`[x]` ✅	v1 approved and downloaded to `docs/forge/intent.md` (SHA-256 `8510138776e8…94cc`)
B2.5	PRD-Spec artifact exported	`[x]` ✅	v1 generated (~95% confidence), approved and downloaded to `docs/forge/prd.md` (SHA-256 `8b138ae25790…3453`)
B2.6	Architecture artifact exported	`[x]` ✅	v1 completed (~100% confidence), approved and downloaded to `docs/forge/architecture.md` (SHA-256 `d0be1c535a37…058b`)
B2.7	User Stories / work orders	`[x]` ✅	User Stories v1 (6 epics / 33 proposed stories, ~75% confidence), approved and downloaded to `docs/forge/work-orders.md` (SHA-256 `a78ac60579ae…cd0d2`)
B2.8	Testing artifact exported	`[x]` ✅	Testing v1 — 132 proposed cases (33 functional / 33 smoke / 33 regression / 33 performance), generated and exported (not approved — no Approve action shown) to `docs/forge/testing.md` (SHA-256 `19749260605e…92b3`). Not executed tests; local suites 108 passing
B2.9	Delivery record + hosted app	`[ ]` ⛔	Delivery stage pending; `docs/forge/delivery.md`; `https://bbs-sg-bank-demo.pages.dev`
B2.10	MCP export credential	`[ ]` 🚫	supplied `forge_` token → HTTP 401 `Invalid or revoked token`; signed-in browser downloads work and were used to export Intent v1 / PRD-Spec v1

B3 — Modernization impact (15%)

	Item	Status	Evidence
B3.1	Algorithmic improvement documented (`O(M·A)`→`O(A+M)`)	`[x]` ✅	`02-before-after.md` §3, §6
B3.2	I/O reduction documented (one write/run)	`[x]` ✅	`02-before-after.md` §6
B3.3	Maintainability / operability comparison	`[x]` ✅	`02-before-after.md` §5, §7
B3.4	Measured speedup, disclosed	`[x]` ✅	`scripts/evidence/benchmark-1200x400.txt` (28.6x), `benchmark-5000x500.txt` (128.1x) — synthetic workload, deliberately inefficient legacy design (`02-...` §6)

B4 — Presentation (10%)

	Item	Status	Evidence
B4.1	Polished slide deck	`[x]` ✅	`slides/index.html`, `slides/SLIDES.md`
B4.2	Demo script timed ≤3 min	`[x]` ✅	`05-demo-script.md`
B4.3	Deck rehearsed end-to-end	`[ ]` ⛔	—
B4.4	Demo recorded	`[x]` ✅	`https://bbs-sg-bank-demo.pages.dev/media/bbs-sg-bank-demo.mp4`

B5 — Content (10%)

	Item	Status	Evidence
B5.1	Documentation pack complete	`[x]` ✅	`docs/README.md` + docs 01–07
B5.2	Post per member published	`[ ]` ⛔	`{MEMBER_n_POST_URL}`
B5.3	All placeholder tokens replaced	`[ ]` ⛔	grep for `{` before submitting

C. Pre-submit QA

[ ] grep -rn "{" sforgehack/docs sforgehack/slides shows only intentional code examples.
[ ] Every speedup figure carries the synthetic-workload / deliberately-inefficient-legacy

disclosure and links a scripts/evidence/ transcript.

[ ] No document states a Forge artifact exists without a linked export.
[ ] No document claims Forge generated or ran the modern engine, deployed anything, that

Testing v1 was approved, that the 33 proposed stories / 132 proposed test cases were implemented or run, or that Forge's proposals (safe temp-file replacement, append-only audit, FastAPI, Forge Shipping) were built — Forge authored the planning documents; the engine is our local code and the local suites are 108 passing.

[ ] No document claims the Forge assessment covers the current working-tree COBOL

(its SHA-256 differs from the uploaded copy).

[ ] Every ✅ row resolves to a real file/line/command.
[ ] Hosted URL loads on a clean browser (no auth).
[ ] Demo video is public and ≤3:00.
[ ] Each member's post is public, original, and names their own contribution.
[ ] 23 September 2026 fields updated.

D. Known open items (honest summary)

✅ Parity + speed measured — isolated GnuCOBOL 3.2.0; byte-identical parity, 28.6x

(1200×400) and 128.1x (5000×500). Only open caveat: figures are synthetic and the legacy design is deliberately inefficient — always disclose.

🟡 **Forge Assessment complete; Intent, PRD-Spec, Architecture and User Stories approved

and exported; Testing v1 generated/exported (not approved); Delivery pending — Assessment is done (57/100, 10 findings; raw report not exported, docs/forge/assessment.md is a team transcription) and Intent v1, PRD-Spec v1, Architecture v1 and User Stories v1 (33 proposed stories) are approved and downloaded into docs/forge/, with Testing v1 (132 proposed cases) generated and exported but not approved. The 33 stories and 132 cases are proposals, not implemented features or executed tests (local suites: 108 passing). The supplied MCP forge_ token is HTTP 401, but signed-in browser downloads work. 30% criterion is partly** satisfied.

✅ Public demo deployed — the Cloudflare Worker runs the JavaScript port of the modern engine. The real COBOL terminal requires local GnuCOBOL.
🟡 73-second demo published; individual public posts pending.
🟡 accounts.dat parsing unguarded (low priority; see 02-... §4).

Current status: public interactive Worker, 73-second video, slide deck and public repository are available. Individual public posts, the submission portal URL, team identity details and Forge Delivery are still outstanding. The hosted Worker runs a JavaScript port; the real COBOL terminal is reproducible locally.