agentic-dev 0.2.11 → 0.2.12

package/sdd/03_verify/10_test/README.md

@@ -1,9 +0,0 @@
-# Test Verification
-
-- 반복 사용하는 검증 harness summary를 유지한다.
-- regression verification summary도 같은 section에서 유지한다.
-
-## Current Documents
-
-- [verification_harness.md](verification_harness.md)
-- [regression_verification.md](regression_verification.md)

package/sdd/03_verify/10_test/regression_verification.md

@@ -1,16 +0,0 @@
-# regression verification
-
-## Status
-
-- pass
-
-## Retained Checks
-
-- `sdd/02_plan/10_test/regression_verification.md`가 current regression scope baseline으로 추가됐다.
-- `AGENTS.md`, `.codex/skills/sdd/SKILL.md`, `sdd/99_toolchain/01_automation/README.md`가 direct-only verification 금지와 selected regression surface 기록 규칙을 함께 유지한다.
-- `sdd/03_build/10_test/regression_verification.md`가 template workflow 반영 상태를 current-state로 설명한다.
-- screen exact automation gate가 Playwright인 경우 suite id, toolchain runner command, artifact path를 retained check에 함께 남긴다.
-
-## Residual Risk
-
-- regression scope selection은 아직 자동 selector가 없어서 문서 규칙과 reviewer 판단에 의존한다.

package/sdd/03_verify/10_test/ui_parity/README.md

@@ -1,4 +0,0 @@
-# UI Parity Artifacts
-
-- UI parity reference asset과 proof output의 current durable 위치다.
-- latest-named artifact를 같은 경로에서 덮어써 유지한다.

package/sdd/03_verify/10_test/verification_harness.md

@@ -1,17 +0,0 @@
-# verification harness
-
-## Status
-
-- pass
-
-## Retained Checks
-
-- verify harness는 stable current artifact path를 사용한다.
-- UI parity reference asset과 proof output은 current durable location을 사용한다.
-- screen local exactness는 `python3 sdd/99_toolchain/01_automation/run_playwright_exactness.py --suite <suite-id> --base-url <url>`를 canonical harness entrypoint로 사용한다.
-- regression verification scope는 별도 durable baseline에서 선택하고, harness는 selected surface 중 automation 가능한 slice를 담당한다.
-
-## Residual Risk
-
-- harness CLI default path가 legacy verify folder를 다시 가리키면 drift가 생긴다.
-- regression surface selector 자동화가 없어 shared-impact 판정은 문서와 reviewer 판단에 의존한다.

package/sdd/03_verify/README.md

@@ -1,22 +0,0 @@
-# Verify Governance
-
-## Purpose
-
-- `sdd/03_verify/`는 현재 구현 상태에 대한 retained verification을 durable 문서로 유지하는 루트다.
-- 날짜별 gate/test log를 쌓지 않고, 현재 기준의 pass/fail 상태와 residual risk만 남긴다.
-
-## Canonical Rule
-
-- `03_verify`는 `02_plan`, `03_build`와 같은 section 축을 따른다.
-- feature, screen, architecture, IAC, test surface별 verification summary를 같은 파일에서 갱신한다.
-- history성 gate/test 로그는 `03_verify`에 두지 않는다.
-
-## Sections
-
-- `01_feature/`: feature 범주별 retained verification summary
-- `02_screen/`: screen별 retained verification summary
-- `03_architecture/`: 거버넌스/구조 관련 verification summary
-- `06_iac/`: delivery/runtime verification summary
-- `07_integration/`: integration verification summary
-- `08_nonfunctional/`: nonfunctional verification summary
-- `10_test/`: 반복 사용하는 검증 harness summary

package/sdd/05_operate/01_runbooks/.gitkeep

@@ -1 +0,0 @@
-

package/sdd/05_operate/01_runbooks/README.md

@@ -1,4 +0,0 @@
-# Runbooks
-
-- 이 폴더는 현재 유효한 운영 절차를 durable runbook으로 유지한다.
-- 날짜별 운영 이력은 두지 않고, 지금 따라야 하는 절차만 갱신한다.

package/sdd/05_operate/02_delivery_status/README.md

@@ -1,4 +0,0 @@
-# Delivery Status
-
-- 이 폴더는 현재 live deployment와 monitoring baseline을 durable 문서로 유지한다.
-- service surface별 current status만 남기고, 과거 release log는 보관하지 않는다.

package/sdd/05_operate/02_delivery_status/service_status.md

@@ -1,16 +0,0 @@
-# service status
-
-## Current Live State
-
-- template repo는 root compose/developer baseline 기준의 current buildable state를 유지한다.
-- production release timeline은 SDD에 누적하지 않는다.
-
-## Monitoring Baseline
-
-- backend `pytest`
-- frontend app build
-- parity harness current output path 확인
-
-## Residual Risk
-
-- 실제 서비스 레포에 이식할 때는 각 서비스의 runtime/edge 기준으로 다시 구체화해야 한다.

package/sdd/05_operate/README.md

@@ -1,12 +0,0 @@
-# Operate Governance
-
-## Purpose
-
-- `sdd/05_operate/`는 현재 배포 상태, 모니터링 기준, 운영 절차를 durable 문서로 유지하는 루트다.
-- raw 운영 로그나 히스토리성 release note는 남기지 않는다.
-
-## Canonical Rule
-
-- runbook은 절차를 설명하는 durable 문서로 유지한다.
-- delivery status는 현재 live state와 확인 기준만 남기고 덮어쓴다.
-- raw runtime log와 상세 incident timeline은 backend/application logging system의 역할이다.

package/sdd/99_toolchain/01_automation/.gitkeep

@@ -1 +0,0 @@
-

package/sdd/99_toolchain/01_automation/README.md

@@ -1,76 +0,0 @@
-# Automation Toolchain
-
-`01_automation`은 screen spec 생성, capture asset 수집, parity harness 실행, Playwright exactness orchestration, recipe 기반 정적 자산 생성을 위한 자동화 자산을 둔다.
-
-## Scope
-
-- screen spec capture script
-- screen spec PDF generator
-- parity harness
-- Playwright exactness runner / suite registry
-- generator manifest
-- asset recipe builder
-- regression verification scope selection rule
-
-## Screen Spec Rule
-
-- feature code: `DOMAIN-FNNN`
-- screen code: `SERVICE-SNNN`
-- desktop capture 기준 visible area: `1710x951`
-- 기본 capture viewport: `1690x940`
-- capture mode: `viewport-first`
-
-## Canonical Tools
-
-- [screen_spec_manifest.py](./screen_spec_manifest.py)
-- [capture_screen_assets.mjs](./capture_screen_assets.mjs)
-- [build_screen_spec_pdf.py](./build_screen_spec_pdf.py)
-- [spec_asset_builder.py](./spec_asset_builder.py)
-- [build_asset_recipes.py](./build_asset_recipes.py)
-- [playwright_exactness_manifest.py](./playwright_exactness_manifest.py)
-- [run_playwright_exactness.py](./run_playwright_exactness.py)
-
-## Storage Rule
-
-- generated runtime asset은 각 구현 repo 경로에 둔다.
-- planning 산출물은 `sdd/01_planning/02_screen/assets/` 아래에 둔다.
-- generator와 manifest는 `sdd/99_toolchain/01_automation/`에 둔다.
-- Playwright harness source는 repo 상황에 따라 `research/agent-browser/pocs/playwright-dev-e2e/` 같은 runtime location에 둘 수 있지만, canonical invocation과 suite registry는 `sdd/99_toolchain/01_automation/`이 소유한다.
-
-## Regression Verification Rule
-
-- toolchain은 builder/harness를 제공하더라도 direct target만 확인하고 종료하는 흐름을 허용하지 않는다.
-- `sdd` 작업은 `sdd/02_plan/10_test/regression_verification.md`를 기준으로 direct, upstream, downstream, shared surface를 선택한다.
-- parity harness나 build check는 회귀 검수의 일부일 뿐이고, 실제 완료 기준은 선택한 regression surface 전체에 대한 retained evidence다.
-- Playwright exactness suite가 있는 surface는 `run_playwright_exactness.py`를 canonical local gate로 사용한다.
-- shared route, shell, auth/session, shared component, API/data contract, generated asset, builder output 변경은 adjacent consumer까지 검수 범위를 넓힌다.
-- 아직 자동화가 없는 회귀 surface는 command/manual verification으로 메우고, automation gap은 `sdd/03_build`, `sdd/03_verify`에 residual risk로 남긴다.
-- Browser Use나 수동 시각 점검은 Playwright exactness gate를 대체하지 않고 보강/진단 용도로만 사용한다.
-
-## Asset Recipe Rule
-
-- 정적 디자인 자산 추출의 canonical tool name은 `스펙에셋빌더`다.
-- generic 실행 파일은 `[spec_asset_builder.py](./spec_asset_builder.py)`다.
-- `[build_asset_recipes.py](./build_asset_recipes.py)`는 기존 호출 호환용 wrapper다.
-- 화면명세서에 있는 icon, logo, illustration, 기타 재사용 가능한 static asset은 먼저 `스펙에셋빌더`로 추출한 뒤 구현 코드에서 사용한다.
-- reusable asset planning 문서는 `sdd/01_planning/02_screen/assets/` 아래에 둔다.
-- 프로젝트별 규칙은 Python manifest에서 `ASSET_RECIPES` 리스트로 선언한다.
-- 각 recipe는 `source`, `crop_box`, `transparent_white_threshold`, `trim`, `output`, `children`을 조합해 재사용 가능한 SVG/PNG 자산을 생성한다.
-- `--verify-exact`를 사용하면 generated output이 source crop과 픽셀 단위로 동일한지 검증한다.
-- builder로 표현 가능한 자산을 수동 redraw나 screenshot crop으로 대체하지 않는다.
-- 수동 예외가 필요하면 plan/build/verify 문서에 source, 예외 사유, 최종 자산 경로를 함께 남긴다.
-- 예시 manifest는 [`../03_templates/asset_recipe_manifest.example.py`](../03_templates/asset_recipe_manifest.example.py)에 둔다.
-- 예시 generated output은 [`../03_templates/generated_assets/README.md`](../03_templates/generated_assets/README.md) 경로를 사용한다.
-
-## Playwright Exactness Rule
-
-- Playwright exactness의 canonical registry는 [playwright_exactness_manifest.py](./playwright_exactness_manifest.py)다.
-- canonical runner는 [run_playwright_exactness.py](./run_playwright_exactness.py)다.
-- downstream repo는 suite source를 runtime location에 둘 수 있지만, retained command는 wrapper 기준으로 남긴다.
-- suite id, spec file, target screen batch, artifact path는 durable current-state로 관리한다.
-- screen 작업의 기본 local exactness gate는 다음 순서를 따른다.
-  - design guide / asset / screen spec baseline 확인
-  - `python3 sdd/99_toolchain/01_automation/run_playwright_exactness.py --suite <suite-id> --base-url <url>`
-  - 필요하면 `--api-base-url`, `--browser`, `--grep`를 추가한다.
-  - 실행 결과와 artifact path를 `sdd/03_verify` current summary에 기록한다.
-- 시작점 예시는 [`../03_templates/playwright_exactness_manifest.example.py`](../03_templates/playwright_exactness_manifest.example.py)를 따른다.

package/sdd/99_toolchain/01_automation/agentic-dev/analyze_proof_results.py

@@ -1,132 +0,0 @@
-#!/usr/bin/env python3
-import json
-import sys
-from pathlib import Path
-
-
-VALID_OK_STATUSES = {"passed"}
-VALID_FAIL_STATUSES = {"failed", "size_mismatch", "missing_reference", "capture_error"}
-
-
-def collect_cases(node, bucket):
-    if isinstance(node, dict):
-        score = None
-        identifier = None
-        for key in ("diff_ratio", "score", "difference_ratio"):
-            value = node.get(key)
-            if isinstance(value, (int, float)):
-                score = float(value)
-                break
-        for key in ("screen_code", "id", "name", "case", "route"):
-            value = node.get(key)
-            if isinstance(value, str) and value.strip():
-                identifier = value.strip()
-                break
-        if score is not None:
-            bucket.append((identifier or f"case_{len(bucket)+1}", score))
-        for value in node.values():
-            collect_cases(value, bucket)
-    elif isinstance(node, list):
-        for item in node:
-            collect_cases(item, bucket)
-
-
-def summarize(path: Path) -> int:
-    data = json.loads(path.read_text(encoding="utf-8"))
-    cases = []
-    collect_cases(data, cases)
-    unique = []
-    seen = set()
-    for identifier, score in sorted(cases, key=lambda item: item[1], reverse=True):
-        key = (identifier, score)
-        if key in seen:
-            continue
-        seen.add(key)
-        unique.append((identifier, score))
-
-    print(f"proof_file={path}")
-    print(f"cases_found={len(unique)}")
-    if not unique:
-        print("No comparable cases with numeric diff_ratio/score found.")
-        return 0
-    failing = [item for item in unique if item[1] > 0]
-    best = min(unique, key=lambda item: item[1])
-    worst = max(unique, key=lambda item: item[1])
-    print(f"failing_cases={len(failing)}")
-    print(f"best_case={best[0]} score={best[1]:.8f}")
-    print(f"worst_case={worst[0]} score={worst[1]:.8f}")
-    for identifier, score in failing[:10]:
-        print(f"- {identifier}: {score:.8f}")
-    return 0
-
-
-def check_gate(path: Path) -> int:
-    data = json.loads(path.read_text(encoding="utf-8"))
-    summary = data.get("summary")
-    screens = data.get("screens")
-    if not isinstance(summary, dict):
-        print("proof_gate=fail reason=missing_summary")
-        return 2
-    if not isinstance(screens, list):
-        print("proof_gate=fail reason=missing_screens")
-        return 2
-
-    failing_statuses = []
-    for screen in screens:
-        if not isinstance(screen, dict):
-            failing_statuses.append(("unknown", "invalid_row"))
-            continue
-        status = screen.get("status")
-        identifier = screen.get("id") or screen.get("route") or "unknown"
-        if status in VALID_OK_STATUSES:
-            continue
-        if status in VALID_FAIL_STATUSES:
-            failing_statuses.append((identifier, status))
-        else:
-            failing_statuses.append((identifier, f"unknown_status:{status}"))
-
-    capture_error = int(summary.get("capture_error", 0) or 0)
-    failed = int(summary.get("failed", 0) or 0)
-    missing_reference = int(summary.get("missing_reference", 0) or 0)
-    matched = bool(summary.get("matched", False))
-
-    if failing_statuses or capture_error > 0 or failed > 0 or missing_reference > 0 or not matched:
-        print("proof_gate=fail")
-        print(f"summary_failed={failed}")
-        print(f"summary_missing_reference={missing_reference}")
-        print(f"summary_capture_error={capture_error}")
-        print(f"summary_matched={matched}")
-        for identifier, status in failing_statuses[:20]:
-            print(f"- {identifier}: {status}")
-        return 2
-
-    print("proof_gate=pass")
-    return 0
-
-
-def main() -> int:
-    if len(sys.argv) < 2:
-        print("Usage: analyze_proof_results.py [--gate] <proof-json>", file=sys.stderr)
-        return 1
-
-    gate_mode = False
-    args = sys.argv[1:]
-    if args and args[0] == "--gate":
-        gate_mode = True
-        args = args[1:]
-    if not args:
-        print("Usage: analyze_proof_results.py [--gate] <proof-json>", file=sys.stderr)
-        return 1
-
-    path = Path(args[0])
-    if not path.is_file():
-        print(f"File not found: {path}", file=sys.stderr)
-        return 1
-
-    if gate_mode:
-        return check_gate(path)
-    return summarize(path)
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())

package/sdd/99_toolchain/01_automation/agentic-dev/analyze_route_gap.py

@@ -1,85 +0,0 @@
-#!/usr/bin/env python3
-import json
-import sys
-from pathlib import Path
-
-
-FAIL_STATUSES = {"missing", "stateful"}
-
-
-def summarize(path: Path) -> int:
-    data = json.loads(path.read_text(encoding="utf-8"))
-    summary = data.get("summary", {})
-    screens = data.get("screens", [])
-    print(f"route_gap_file={path}")
-    print(f"service={data.get('service', 'unknown')}")
-    print(f"total={summary.get('total', 0)}")
-    print(f"direct={summary.get('direct', 0)}")
-    print(f"shared={summary.get('shared', 0)}")
-    print(f"stateful={summary.get('stateful', 0)}")
-    print(f"missing={summary.get('missing', 0)}")
-    for row in screens:
-        if isinstance(row, dict) and row.get("status") in FAIL_STATUSES:
-            print(f"- {row.get('id', 'unknown')}: {row.get('status')}")
-    return 0
-
-
-def check_gate(path: Path) -> int:
-    data = json.loads(path.read_text(encoding="utf-8"))
-    summary = data.get("summary")
-    screens = data.get("screens")
-    if not isinstance(summary, dict):
-        print("route_gap_gate=fail reason=missing_summary")
-        return 2
-    if not isinstance(screens, list):
-        print("route_gap_gate=fail reason=missing_screens")
-        return 2
-
-    failing_rows = []
-    for row in screens:
-        if not isinstance(row, dict):
-            failing_rows.append(("unknown", "invalid_row"))
-            continue
-        status = str(row.get("status", "")).strip().lower()
-        identifier = row.get("id") or row.get("title") or "unknown"
-        if status in FAIL_STATUSES:
-            failing_rows.append((identifier, status))
-
-    if failing_rows or int(summary.get("missing", 0) or 0) > 0 or int(summary.get("stateful", 0) or 0) > 0:
-        print("route_gap_gate=fail")
-        print(f"summary_missing={summary.get('missing', 0)}")
-        print(f"summary_stateful={summary.get('stateful', 0)}")
-        for identifier, status in failing_rows[:20]:
-            print(f"- {identifier}: {status}")
-        return 2
-
-    print("route_gap_gate=pass")
-    return 0
-
-
-def main() -> int:
-    if len(sys.argv) < 2:
-        print("Usage: analyze_route_gap.py [--gate] <route-gap-json>", file=sys.stderr)
-        return 1
-
-    gate_mode = False
-    args = sys.argv[1:]
-    if args and args[0] == "--gate":
-        gate_mode = True
-        args = args[1:]
-    if not args:
-        print("Usage: analyze_route_gap.py [--gate] <route-gap-json>", file=sys.stderr)
-        return 1
-
-    path = Path(args[0])
-    if not path.is_file():
-        print(f"File not found: {path}", file=sys.stderr)
-        return 1
-
-    if gate_mode:
-        return check_gate(path)
-    return summarize(path)
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())

package/sdd/99_toolchain/01_automation/agentic-dev/assets/repo-contract.template.json

@@ -1,75 +0,0 @@
-{
-  "schema_version": 1,
-  "name": "replace-service-name",
-  "spec": {
-    "planning_paths": [
-      "sdd/01_planning",
-      "sdd/02_plan"
-    ],
-    "canonical_targets_dir": "sdd/02_plan"
-  },
-  "frontend": {
-    "default_target": "web",
-    "targets": {
-      "web": {
-        "dir": "client/web",
-        "adapter_path": "client/web/scripts/ui-parity-web-adapter.mjs",
-        "screens_path": "client/web/src/lib/specScreens.json",
-        "routes_path": "client/web/src/lib/specRouteCatalog.json",
-        "preview_url": "http://127.0.0.1:4301",
-        "parity_contract_path": "sdd/02_plan/10_test/templates/ui_parity_web_contract.yaml",
-        "route_gap_output": "sdd/02_plan/99_generated/from_planning/ui_parity/ui_parity_web_route_gap_report.json",
-        "route_gap_markdown_output": "sdd/02_plan/99_generated/from_planning/ui_parity/ui_parity_web_route_gap_report.md",
-        "proof_output": "sdd/03_verify/10_test/ui_parity/templates_web_agentic_dev_latest.json"
-      },
-      "mobile": {
-        "dir": "client/mobile",
-        "adapter_path": "client/mobile/scripts/ui-parity-mobile-adapter.mjs",
-        "screens_path": "client/mobile/src/lib/specScreens.json",
-        "routes_path": "client/mobile/src/lib/specRouteCatalog.json",
-        "preview_url": "http://127.0.0.1:4302",
-        "parity_contract_path": "sdd/02_plan/10_test/templates/ui_parity_mobile_contract.yaml",
-        "route_gap_output": "sdd/02_plan/99_generated/from_planning/ui_parity/ui_parity_mobile_route_gap_report.json",
-        "route_gap_markdown_output": "sdd/02_plan/99_generated/from_planning/ui_parity/ui_parity_mobile_route_gap_report.md",
-        "proof_output": "sdd/03_verify/10_test/ui_parity/templates_mobile_agentic_dev_latest.json"
-      },
-      "admin": {
-        "dir": "client/admin",
-        "adapter_path": "client/admin/scripts/ui-parity-admin-adapter.mjs",
-        "screens_path": "client/admin/src/lib/specScreens.json",
-        "routes_path": "client/admin/src/lib/specRouteCatalog.json",
-        "preview_url": "http://127.0.0.1:4400",
-        "parity_contract_path": "sdd/02_plan/10_test/templates/ui_parity_admin_contract.yaml",
-        "route_gap_output": "sdd/02_plan/99_generated/from_planning/ui_parity/ui_parity_admin_route_gap_report.json",
-        "route_gap_markdown_output": "sdd/02_plan/99_generated/from_planning/ui_parity/ui_parity_admin_route_gap_report.md",
-        "proof_output": "sdd/03_verify/10_test/ui_parity/templates_admin_agentic_dev_latest.json"
-      },
-      "landing": {
-        "dir": "client/landing",
-        "adapter_path": "client/landing/scripts/ui-parity-landing-adapter.mjs",
-        "screens_path": "client/landing/src/lib/specScreens.json",
-        "routes_path": "client/landing/src/lib/specRouteCatalog.json",
-        "preview_url": "http://127.0.0.1:4300",
-        "parity_contract_path": "sdd/02_plan/10_test/templates/ui_parity_landing_contract.yaml",
-        "route_gap_output": "sdd/02_plan/99_generated/from_planning/ui_parity/ui_parity_landing_route_gap_report.json",
-        "route_gap_markdown_output": "sdd/02_plan/99_generated/from_planning/ui_parity/ui_parity_landing_route_gap_report.md",
-        "proof_output": "sdd/03_verify/10_test/ui_parity/templates_landing_agentic_dev_latest.json"
-      }
-    }
-  },
-  "commands": {
-    "plan_audit": "bash sdd/99_toolchain/01_automation/agentic-dev/run_frontend_target.sh route_gap .",
-    "build": "bash sdd/99_toolchain/01_automation/agentic-dev/run_frontend_target.sh build .",
-    "proof": "bash sdd/99_toolchain/01_automation/agentic-dev/run_frontend_target.sh proof .",
-    "deploy_dev": "echo 'replace deploy_dev command' >&2",
-    "verify_dev": "echo 'replace verify_dev command' >&2"
-  },
-  "artifacts": {
-    "route_gap_output": "sdd/02_plan/99_generated/from_planning/ui_parity/ui_parity_web_route_gap_report.json",
-    "proof_output": "sdd/03_verify/10_test/ui_parity/replace_agentic_dev_latest.json"
-  },
-  "notes": [
-    "Replace the commands and proof output path after cloning the template.",
-    "plan_audit is expected to generate and gate a route-gap report before proof."
-  ]
-}

package/sdd/99_toolchain/01_automation/agentic-dev/bootstrap_frontend_parity.sh

@@ -1,84 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-
-usage() {
-  cat <<'EOF'
-Usage: bootstrap_frontend_parity.sh [repo_root] [target]
-
-Run first-time frontend parity bootstrap for a cloned template.
-
-Steps:
-  1. init repo contract and route-gap assets
-  2. build frontend
-  3. start preview server
-  4. materialize reference assets
-  5. run plan_audit gate
-  6. run proof gate
-EOF
-}
-
-if [[ "${1:-}" == "--help" || "${1:-}" == "-h" ]]; then
-  usage
-  exit 0
-fi
-
-repo_root="$(cd "${1:-$(pwd)}" && pwd)"
-frontend_target="${2:-}"
-script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-target_payload="$(python3 "${script_dir}/resolve_frontend_target.py" "${repo_root}" "${frontend_target}")"
-read_target_value() {
-  local key="$1"
-  python3 - "${target_payload}" "${key}" <<'PY'
-import json
-import sys
-
-payload = json.loads(sys.argv[1])
-key = sys.argv[2]
-value = payload
-for part in key.split("."):
-    if not isinstance(value, dict):
-        raise SystemExit(1)
-    value = value.get(part)
-if value is None:
-    raise SystemExit(1)
-print(value if isinstance(value, str) else json.dumps(value, ensure_ascii=False))
-PY
-}
-target_name="$(read_target_value "name")"
-target_dir="$(read_target_value "target.dir")"
-preview_url="$(read_target_value "target.preview_url")"
-preview_log="${repo_root}/sdd/03_verify/10_test/ui_parity/${target_name}-preview.log"
-preview_pid=""
-
-cleanup() {
-  if [[ -n "${preview_pid}" ]] && kill -0 "${preview_pid}" >/dev/null 2>&1; then
-    kill "${preview_pid}" >/dev/null 2>&1 || true
-    wait "${preview_pid}" >/dev/null 2>&1 || true
-  fi
-}
-trap cleanup EXIT
-
-bash "${script_dir}/init_frontend_parity.sh" "${repo_root}" "${frontend_target}"
-
-cd "${repo_root}"
-mkdir -p "$(dirname "${preview_log}")"
-bash "${script_dir}/run_frontend_target.sh" build "${repo_root}" "${frontend_target}"
-npm --prefix "${target_dir}" run preview >"${preview_log}" 2>&1 &
-preview_pid="$!"
-for _ in $(seq 1 30); do
-  if curl -fsS "${preview_url}" >/dev/null 2>&1; then
-    break
-  fi
-  sleep 1
-done
-if ! curl -fsS "${preview_url}" >/dev/null 2>&1; then
-  echo "Preview server did not become ready: ${preview_url}" >&2
-  echo "Preview log: ${preview_log}" >&2
-  exit 1
-fi
-bash "${script_dir}/run_frontend_target.sh" materialize_references "${repo_root}" "${frontend_target}"
-
-bash "${script_dir}/run_repo_phase.sh" plan_audit "${repo_root}" "${frontend_target}"
-bash "${script_dir}/run_repo_phase.sh" proof "${repo_root}" "${frontend_target}"
-
-echo "frontend_parity_bootstrap=ok"

package/sdd/99_toolchain/01_automation/agentic-dev/init_frontend_parity.sh

@@ -1,33 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-
-usage() {
-  cat <<'EOF'
-Usage: init_frontend_parity.sh [repo_root] [target]
-
-Bootstrap repo-local frontend parity assets for a cloned template.
-
-Steps:
-  1. Ensure repo contract aliases exist
-  2. Scaffold ui parity contract
-  3. Generate route-gap json/markdown manifests
-
-Optional runtime steps such as reference materialization and proof are intentionally excluded.
-EOF
-}
-
-if [[ "${1:-}" == "--help" || "${1:-}" == "-h" ]]; then
-  usage
-  exit 0
-fi
-
-repo_root="$(cd "${1:-$(pwd)}" && pwd)"
-frontend_target="${2:-}"
-script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-
-bash "${script_dir}/init_repo_contract.sh" "${repo_root}" >/dev/null
-
-bash "${script_dir}/run_frontend_target.sh" scaffold "${repo_root}" "${frontend_target}"
-bash "${script_dir}/run_frontend_target.sh" route_gap "${repo_root}" "${frontend_target}"
-
-echo "frontend_parity_init=ok"