出具缺口不再把节点类型说两遍：上游缺口的前缀在节点名就是类型叫法时只写一次（「反方审稿：」），双盲复算、反方审稿的缺口正文去掉自带的类型（「推翻了 1 句结论：…」「指标「X」两路结果不一致，审核人选择了盲算结果」）；时间线和报错里那句照旧带主语

backend/app/engine/nodes/challenge.py

@@ -466,9 +466,15 @@ def _named(claims: list[dict[str, Any]]) -> str:
     return names + ("等" if len(claims) > _NAMED else "")
 
 
+def _refuted_gap(claims: list[dict[str, Any]]) -> str:
+    """出具缺口里的那句：「推翻了 N 句结论：「…」」。不带节点名也不带「反方审稿」：出具契约统一加「反方审稿「节点名」：」
+    （没改名的节点只加「反方审稿：」），正文再说一遍就成了「反方审稿：反方审稿推翻了……」。"""
+    return f"推翻了 {len(claims)} 句结论：{_named(claims)}"
+
+
 def _refuted_words(claims: list[dict[str, Any]]) -> str:
-    """出具缺口、时间线、报错里的那句：「反方审稿推翻了 N 句结论：「…」」。不带节点名：出具契约统一加前缀。"""
-    return f"反方审稿推翻了 {len(claims)} 句结论：{_named(claims)}"
+    """时间线、报错里的那句：「反方审稿推翻了 N 句结论：「…」」。它们不经出具契约加前缀，主语得自己带上。"""
+    return f"反方审稿{_refuted_gap(claims)}"
 
 
 def pick_items(claims: list[dict[str, Any]]) -> list[dict[str, Any]]:
@@ -676,7 +682,7 @@ def _finish(ctx: NodeContext, state: GraphState, outcome: dict[str, Any], *, on_
     result: dict[str, Any] = {"challenge_artifact": outcome.get("artifact"), "counts": outcome.get("counts") or {},
                               "items": pick_items(claims), "on_refuted": on_refuted}
     if refuted and on_refuted == "degrade":
-        result["issuance_gaps"] = [_refuted_words(refuted)]
+        result["issuance_gaps"] = [_refuted_gap(refuted)]
     updates: dict[str, Any] = {"nodes": {ctx.node.id: result}}
     spent = outcome.get("usage") or {}
     if spent.get("calls"):

backend/app/engine/nodes/dual_check.py

@@ -110,10 +110,11 @@ SAME_SQL_ASPECT = "data"
 DIFF_TIMEOUT = "两条 SQL 比较超时，未能归纳成具体差异，请对照原文"
 DIFF_FAILED = "两条 SQL 比较时出错，未能归纳成具体差异，请对照原文"
 NO_SQL = "有一边没有可比较的 SQL，请对照原文"
-#: 交给成果节点的缺口（不带节点名：出具契约统一加「双盲复算「节点名」：」）
-GAP_BLIND = "指标「{name}」双盲复算不一致，审核人选择了盲算结果"
-GAP_NEITHER = "指标「{name}」双盲复算不一致，审核人判断两边都不对"
-GAP_UNANSWERED = "指标「{name}」双盲复算不一致，审核人没有作出选择"
+#: 交给成果节点的缺口。不带节点名，也不再说「双盲复算」：出具契约统一加「双盲复算「节点名」：」（没改名的节点只加
+#: 「双盲复算：」），正文再说一遍就成了「双盲复算：指标「X」双盲复算不一致……」
+GAP_BLIND = "指标「{name}」两路结果不一致，审核人选择了盲算结果"
+GAP_NEITHER = "指标「{name}」两路结果不一致，审核人判断两边都不对"
+GAP_UNANSWERED = "指标「{name}」两路结果不一致，审核人没有作出选择"
 GAP_UNCHECKED = "指标「{name}」无法复算：{reason}"
 
 BLIND_SYSTEM = (

backend/app/engine/nodes/io.py

@@ -322,7 +322,8 @@ def _upstream_issuance_gaps(state: GraphState, ctx: NodeContext) -> list[str]:
     这是给「把关类」节点留的通用口子：节点觉得这次的结论不该完整出具（双盲复算两路对不上、
     反方审稿的反驳没有回应），在自己的产出里写 issuance_gaps，成果节点的出具契约照单收进缺口，
     出具随之降档。写的时候不用带节点名，这里统一加上「双盲复算「复核销售额」：」这样的前缀，
-    和别的缺口一样读得出是谁报的。
+    和别的缺口一样读得出是谁报的。节点名就是类型的叫法（拖进画布没改名）时只写一次：「反方审稿：」，
+    不写成「反方审稿「反方审稿」：」。缺口正文也不必再提节点类型，前缀已经说了。
 
     只认这个成果节点的上游：另一条支路上的节点管不到这份成果。不是字符串、空白的条目跳过——
     这个键谁都能写（数据整形的 JSON 模板也行），写坏了只当没写，不能让出具崩掉。
@@ -340,7 +341,8 @@ def _upstream_issuance_gaps(state: GraphState, ctx: NodeContext) -> list[str]:
         reported = payload.get("issuance_gaps")
         if not isinstance(reported, list):
             continue
-        prefix = f"{type_label(node.type)}「{node.title}」："
+        kind = type_label(node.type)
+        prefix = f"{kind}：" if node.title == kind else f"{kind}「{node.title}」："
         out.extend(prefix + gap.strip() for gap in reported if isinstance(gap, str) and gap.strip())
     return out
 

backend/tests/test_challenge_node.py

@@ -265,7 +265,7 @@ async def test_a_refuted_conclusion_degrades_the_issuance(monkeypatch, shop):
 
     issuance = row.output["_issuance"]
     assert issuance["tier"] == "degraded"
-    assert issuance["gaps"] == [f"反方审稿「审稿」：反方审稿推翻了 1 句结论：「{SOUTH}」"], "缺口不带节点名，前缀由出具契约加"
+    assert issuance["gaps"] == [f"反方审稿「审稿」：推翻了 1 句结论：「{SOUTH}」"], "缺口不带节点名，前缀由出具契约加"
 
     doc = await report_doc(row.id)
     east, south = unit_of(doc, EAST), unit_of(doc, SOUTH)
@@ -275,7 +275,7 @@ async def test_a_refuted_conclusion_degrades_the_issuance(monkeypatch, shop):
     assert out["items"] == [{"unit": east, "text": EAST, "verdict": "weakened"},
                             {"unit": south, "text": SOUTH, "verdict": "refuted"}]
     assert out["on_refuted"] == "degrade"
-    assert out["issuance_gaps"] == [f"反方审稿推翻了 1 句结论：「{SOUTH}」"]
+    assert out["issuance_gaps"] == [f"推翻了 1 句结论：「{SOUTH}」"]
 
     record = artifact_store.load(out["challenge_artifact"])
     assert set(record) == {"report", "doc_artifact", "claims", "model"}
@@ -317,6 +317,19 @@ async def test_a_refuted_conclusion_degrades_the_issuance(monkeypatch, shop):
     assert finished.seq <= row.manifest_seq
 
 
+async def test_a_challenge_named_after_its_type_is_named_once_in_the_gap(monkeypatch, shop):
+    """拖进画布没改名的反方审稿：缺口读成「反方审稿：推翻了 1 句结论：…」，前缀和正文都不再重复「反方审稿」"""
+    Models(monkeypatch, shop)
+    spec = graph(shop)
+    next(n for n in spec["nodes"] if n["id"] == "chal")["data"]["label"] = "反方审稿"
+    row = await finish(spec)
+    assert row.status == "succeeded", row.error
+    assert row.output["_issuance"]["gaps"] == [f"反方审稿：推翻了 1 句结论：「{SOUTH}」"]
+    # 时间线、报错里那一句照旧带主语：它们不经出具契约加前缀
+    [refuted] = await logs(row.id, "challenge_refuted")
+    assert refuted["message"] == f"反方审稿推翻了 1 句结论：「{SOUTH}」"
+
+
 async def test_flag_only_records_and_fail_stops_the_run(monkeypatch, shop):
     Models(monkeypatch, shop)
     row = await finish(graph(shop, on_refuted="flag"))
@@ -843,7 +856,7 @@ async def test_a_fail_that_would_be_swallowed_degrades_instead(monkeypatch, shop
     row = await finish(graph(shop, on_refuted="fail", on_error="continue"))
     assert row.status == "succeeded", row.error
     out = await output_of(row.id, "chal")
-    assert out["on_refuted"] == "degrade" and out["issuance_gaps"] == [f"反方审稿推翻了 1 句结论：「{SOUTH}」"]
+    assert out["on_refuted"] == "degrade" and out["issuance_gaps"] == [f"推翻了 1 句结论：「{SOUTH}」"]
     assert row.output["_issuance"]["tier"] == "degraded"
 
 

backend/tests/test_dual_check_node.py

@@ -446,13 +446,25 @@ async def test_choosing_the_blind_result_degrades_and_writes_the_catalog(monkeyp
     done = await decide(client, run.id, {"gmv": "b"})
     assert done.status == "succeeded", done.error
     out = dual_output(done, await events(run.id))
-    assert out["issuance_gaps"] == ["指标「销售额」双盲复算不一致，审核人选择了盲算结果"]
+    assert out["issuance_gaps"] == ["指标「销售额」两路结果不一致，审核人选择了盲算结果"]
     issuance = done.output["_issuance"]
     assert issuance["tier"] == "degraded"
-    assert issuance["gaps"] == ["双盲复算「复核销售额」：指标「销售额」双盲复算不一致，审核人选择了盲算结果"]
+    assert issuance["gaps"] == ["双盲复算「复核销售额」：指标「销售额」两路结果不一致，审核人选择了盲算结果"]
     assert (await notes_of(source_id)).notes["business_date"]["value"] == {"column": "paid_date"}
 
 
+async def test_a_dual_check_named_after_its_type_is_named_once_in_the_gap(monkeypatch, shop, client):
+    """拖进画布没改名的双盲复算：缺口读成「双盲复算：指标「销售额」两路结果不一致……」，不写两遍「双盲复算」"""
+    name, _ = shop
+    Blind(monkeypatch, name, {"销售额": (RIGHT, "gmv")})
+    spec = graph(name, BY_ORDER_DATE)
+    next(n for n in spec["nodes"] if n["id"] == "dual")["data"]["label"] = "双盲复算"
+    run = await start(spec)
+    done = await decide(client, run.id, {"gmv": "neither"})
+    assert done.status == "succeeded", done.error
+    assert done.output["_issuance"]["gaps"] == ["双盲复算：指标「销售额」两路结果不一致，审核人判断两边都不对"]
+
+
 async def test_neither_degrades_and_writes_nothing(monkeypatch, shop, client):
     name, source_id = shop
     Blind(monkeypatch, name, {"销售额": (RIGHT, "gmv")})
@@ -461,7 +473,7 @@ async def test_neither_degrades_and_writes_nothing(monkeypatch, shop, client):
     assert done.status == "succeeded", done.error
     evs = await events(run.id)
     out = dual_output(done, evs)
-    assert out["issuance_gaps"] == ["指标「销售额」双盲复算不一致，审核人判断两边都不对"]
+    assert out["issuance_gaps"] == ["指标「销售额」两路结果不一致，审核人判断两边都不对"]
     assert done.output["_issuance"]["tier"] == "degraded"
     assert not logs(evs, "dual_check_catalog") and await notes_of(source_id) is None
 
@@ -476,7 +488,7 @@ async def test_a_missing_choice_is_recorded_as_no_choice(monkeypatch, shop, clie
     evs = await events(run.id)
     out = dual_output(done, evs)
     assert out["choices"] == {} and "choice" not in out["metrics"][0]
-    assert out["issuance_gaps"] == ["指标「销售额」双盲复算不一致，审核人没有作出选择"]
+    assert out["issuance_gaps"] == ["指标「销售额」两路结果不一致，审核人没有作出选择"]
     [resolved] = logs(evs, "dual_check_resolved")
     assert resolved["choices"] == {} and resolved["gaps"] == out["issuance_gaps"]
     assert "没有作出选择" in resolved["message"] and "两边都不对" not in resolved["message"]
@@ -505,7 +517,7 @@ async def test_on_reject_fail_stops_the_run(monkeypatch, shop, client):
     assert done.status == "failed" and "审核人没有采用原流程的结果" in (done.error or "") and "销售额" in done.error
     evs = await events(run.id)
     assert [d["gaps"] for d in logs(evs, "dual_check_resolved")] == [
-        ["指标「销售额」双盲复算不一致，审核人选择了盲算结果"]]
+        ["指标「销售额」两路结果不一致，审核人选择了盲算结果"]]
     # 审核人的选择照样写进了目录：那是对数据的判断，和运行停不停无关
     entry = await notes_of(source_id)
     assert entry.notes["business_date"]["value"] == {"column": "paid_date"} and entry.version == 1

backend/tests/test_issuance_upstream_gaps.py

@@ -93,3 +93,9 @@ async def test_without_issuance_gaps_nothing_changes():
     issuance = row.output["_issuance"]
     assert issuance["tier"] == "formal"
     assert issuance["gaps"] == []
+
+
+async def test_a_node_named_after_its_type_is_named_once():
+    """节点名就是类型的叫法（拖进画布没改名的）：前缀只写一次「数据整形：」，不写成「数据整形「数据整形」：」"""
+    row = await finish(graph(gaps_node("recheck", "数据整形", ["两路结果不一致"])))
+    assert row.output["_issuance"]["gaps"] == ["数据整形：两路结果不一致"]