# AWS CardDemo Modernization Final Proof

This bundle packages the local AWS CardDemo modernization run as a reproducible technical-review artifact. It does not add application features; it records the source identity, modern code location, commands, receipts, static-oracle replay, no-cost confidence proof, A-grade promotion gate, known limits, and clean-copy verification status.

## Source Identity

- Source repo: `https://github.com/aws-samples/aws-mainframe-modernization-carddemo.git`
- Source version: `59cc6c2fd7ebd7ef7925cad552a01a4b8b6e4d5e`
- Source local path used for this run: `D:\scraper_storage\modernization_labs\aws-carddemo-modernization-20260523T073000Z\source`
- Source file hash count in receipt: `204`
- Source hash digest over receipt map: `0703f97553bde0f5bd170aa95ec8c771a975b397be7d9e8088ec6fbf31d333c2`

## Modern Code

- Modern code path: `D:\scraper_storage\modernization_labs\aws-carddemo-modernization-20260523T073000Z\modern_carddemo`
- Modern file hash count in receipt: `29`
- Modern hash digest over receipt map: `599294a7df7226c4d2a4b820de4bf567abbd1bdd1e41a1b6fd123457cb3abf40`

## One-Command Setup And Run

Run these from the modern code root with the AWS CardDemo source clone/copy at `..\source`.

Setup database:

```powershell
python -m carddemo.cli init-db --source-root ..\source --db-path .\carddemo.sqlite
```

Run web/API server:

```powershell
python -m carddemo.cli serve --source-root ..\source --db-path .\carddemo.sqlite --port 8087
```

Open `http://127.0.0.1:8087`.

## One-Command Tests

```powershell
python -m unittest discover -s tests
```

## One-Command Static-Oracle Replay

```powershell
python tools\replay_static_oracles.py --source-root ..\source --out-dir .\reports --source-head 59cc6c2fd7ebd7ef7925cad552a01a4b8b6e4d5e
```

## One-Command A-Grade Promotion Gate

```powershell
python tools\evaluate_a_grade_promotion.py --source-root ..\source --out-dir .\final_proof
```

## One-Command No-Cost Confidence Proof

```powershell
python tools\build_no_cost_confidence.py --source-root ..\source --out-dir .\final_proof
```

## Full Local Proof Refresh

```powershell
powershell -NoProfile -Command "python -m carddemo.cli init-db --source-root ..\source --db-path .\carddemo.sqlite; python -m unittest discover -s tests; python tools\generate_receipts.py --source-root ..\source --db-path .\carddemo.sqlite --out-dir .\reports; python tools\replay_static_oracles.py --source-root ..\source --out-dir .\reports --source-head 59cc6c2fd7ebd7ef7925cad552a01a4b8b6e4d5e; python tools\evaluate_a_grade_promotion.py --source-root ..\source --out-dir .\final_proof; python tools\build_no_cost_confidence.py --source-root ..\source --out-dir .\final_proof; python tools\package_final_proof.py --source-root ..\source"
```

## Final Receipt Summary

- Receipt hash: `fea9dfeca8e1b400afe6fbf4a25f92282a6330da0bb031f87b8dbd6e41044819`
- Receipt generated: `2026-05-23T15:01:51+00:00`
- Static replay hash: `912e98e11d46cd967070d06ec17aaed3402bed44026267d1906ea6a2d54b9a42`
- Static replay passed: `True`
- Runtime probe count: `6`
- Failed replay checks: `[]`

## Parity Summary

Under the current static/source-oracle rubric:

- Capability count: `50`
- B-grade static/source-oracle capabilities: `50`
- F-grade gaps: `0`
- Gap count: `0`
- Next-gap ranking count: `0`
- Static-oracle-or-better count: `50`
- A-grade live legacy runtime trace count: `0`

Proof grade meanings are recorded in `FINAL_RECEIPT_SUMMARY.json`.

## No-Cost Confidence Proof

- Reviewer posture: `strong_pre_runtime_confidence`
- Ready for pre-runtime review: `True`
- Paid runtime used: `False`
- Legacy runtime claim made: `False`
- Source-backed static capabilities: `50`
- Source refs in capability ledger: `101`
- Golden traces / edge cases: `34` / `40`
- Remaining gap: `live legacy COBOL/CICS/VSAM/JCL runtime validation on a real or licensed-compatible runtime`
- No-cost confidence report hash: `c72a4b6b30b7bbee9096b3d79e2893fb0263d5aee1f4aebaec37401810208441`

This is the maximum unpaid/pre-runtime evidence layer. It is meant to convince a technical reviewer that the modernization is probably functional under the extracted legacy-derived behavior, while preserving the live runtime validation gap.

## A-Grade Promotion Gate

- A-grade gate decision: `do_not_promote`
- Promotable to A: `False`
- Required live legacy runtime matches: `50`
- Accepted live legacy runtime matches: `0`
- Missing live legacy runtime matches: `50`
- A-grade report hash: `4f962c34517806cf2dee00f3fdd0b7c4f51e9ec3f5935eedfabd04104e3ad3af`

The A-grade gate is intentionally stricter than the B-grade static/source-oracle proof. It requires accepted live legacy COBOL/CICS runtime trace matches for each capability before any capability can be promoted to A.

## Smoke Evidence

- Screenshot: `reports/ui_batch_reports_smoke.png` SHA-256 `f192f2b751ee902b70defbb5b49b745bbf8855a10bb8dc23d60e6c8bb9f6e537`.
- Clean-copy API smoke: passed=True port=8091 endpoints=batch_resource_setup_tranidx, health, mq_system_date, summary.

## Clean-Copy Verification

- Clean-copy verification status: `passed`
- Verification report included in `FINAL_RECEIPT_SUMMARY.json` when present.

## Caveat

This proof is static/source-oracle parity plus executable modern-service replay. It is not live COBOL/CICS all-path runtime equivalence.

See `KNOWN_LIMITS.md` for the full claim boundary.