docs(gdd): 用语规范批 W2——systems 十九篇词面换代 74 处（收口 21/不得 13/报错 7/若干 8/对拍 3/在飞 1/落文 2/随批 2 等按 A 表与语境义改写）+死链修复 2 处（武器锻造→锻造工坊漏改指路）+锻造工坊文头 v0.7→v0.8 失同步补齐；doc_term_audit 增设随批/台账专项补扫通道（白名单自豁免词的版本历史节外可见性）。秘宝充能 2 处与对照性收口 3 处按豁免保留

tools/doc_term_audit.py

@@ -60,6 +60,37 @@ RETIRED_TERMS = [
 
 DEFAULT_DIRS = ["design/gdd", "design/registry"]
 
+# 用语规范 A 表补充词专项补扫：随批/台账 两词本身是上方行级白名单信号（行内出现即归
+# 留档类），主扫描对它们自豁免；本通道只做版本历史节外的专项报告，供后续触碰批清偿。
+SUPPLEMENT_TERMS = [
+    ("随批", "2026-10-15", "顺带", "用语规范 A 表（doc-conventions 第 4 节）"),
+    ("台账", "2026-10-15", "状态记录", "用语规范 A 表（doc-conventions 第 4 节）"),
+]
+
+
+def scan_supplement(paths):
+    terms = [(re.compile(t), date, repl, note) for t, date, repl, note in SUPPLEMENT_TERMS]
+    findings = []
+    for p in paths:
+        text = Path(p).read_text(encoding="utf-8").replace("\r\n", "\n")
+        in_version_section = False
+        for lineno, line in enumerate(text.split("\n"), 1):
+            if SECTION_RE.match(line):
+                in_version_section = True
+            hits = [(m.group(0), t) for t in terms for m in [t[0].search(line)] if m]
+            if not hits or in_version_section or COMMENT_RE.match(line):
+                continue
+            findings.append({
+                "term": "、".join(sorted({h[0] for h in hits})),
+                "file": str(p).replace("\\", "/"),
+                "line": lineno,
+                "klass": "疑似现行",
+                "text": line.strip()[:160],
+                "retired": hits[0][1][1],
+                "replacement": hits[0][1][2],
+            })
+    return findings
+
 # 词级语境豁免：某词退役的只是特定机制语境，另一现行机制合法沿用同一词时，
 # 命中行含豁免语境信号即不算退役词命中（不进疑似、不进留档计数）。
 # 充能：挖掘期/战斗段充能退役（2026-09-19），但秘宝主动件「每冒险充能」（同调件
@@ -134,7 +165,7 @@ def main():
             paths += sorted(Path(d).rglob("*.yaml"))
     paths = [p for p in paths if Path(p).exists()]
 
-    findings = scan(paths)
+    findings = scan(paths) + scan_supplement(paths)
     suspect = [f for f in findings if f["klass"] == "疑似现行"]
     kept = [f for f in findings if f["klass"] != "疑似现行"]