diff --git a/continuation.md b/continuation.md
index 5a4e1da..266cb05 100644
--- a/continuation.md
+++ b/continuation.md
@@ -4,6 +4,29 @@
 
 ---
 
+## 本次完成 (2026-08-22) — P1-1 修复 dedup 高重复率返回码语义
+
+**问题**:`run_dedup` 重复率 > 5% 时返回 1;`scheduler/pipeline.py` 曾把 `dedup` 的 rc=1 无条件视为成功并打日志"无新数据场景"。结果:高重复率(可能是正常无新数据,也可能是抓取源/指纹库异常)被一刀切掩盖,真实异常无法上报。
+
+**修复(分离"执行成功"与"统计告警")**:
+- `scripts/run_dedup.py`:
+  - 生产模式(默认):重复率仅作 **WARNING 告警**,不影响退出码(执行成功即 0);阈值常量 `_DUP_RATE_THRESHOLD = 0.05`
+  - 新增 `--strict` 验收模式:保留 M3 验收门槛,重复率 > 5% 时返回 1(供人工验收)
+  - 新增统计快照 `data/deduped/{date}/stats.json`(原子写):unique/duplicates/total/dup_rate/layers/fingerprint_total/generated_at,供运维排查与监控
+  - 顺带修复:空日场景 `sources.json` 写入前补 `mkdir`(原会 FileNotFoundError)
+- `scheduler/pipeline.py`:移除 `dedup` rc=1 特判,恢复"非 0 即失败"的统一语义
+
+**测试**:
+- 新增 `tests/test_run_dedup.py` 7 个(生产模式高重复返回 0 / --strict 高重复返回 1 / --strict 低重复返回 0 / 空日返回 0 / stats.json 结构与数值 / pipeline 不再掩蔽失败)
+- 全量 **266 passed**,3 个 crawler 基线失败与本次无关
+
+**生产实测**:同日重跑重复率 100% → 生产模式 `rc=0` + WARNING + stats.json 完整;`--strict` 同场景 `rc=1`(用非管道方式核实退出码)
+
+**待办/遗留**:
+- P1-2(守护进程补跑 steps 含 cninfo 三步)、P1-3(调度时区与 date.today() 一致性),待用户决策
+
+---
+
 ## 本次完成 (2026-08-22) — P0-3/P0-4 补跑日期语义修复
 
 **P0-3(crawler 补跑历史日期静默空转)**:
@@ -20,7 +43,7 @@
 - 全量 **259 passed**,3 个 crawler 基线失败与本次无关
 
 **待办/遗留**:
-- P1:dedup rc=1 静默转成功(天天重复率>5% 被掩蔽)、守护进程补跑 steps 含 cninfo_*、调度时区与 date.today() 一致性,待用户决策
+- ~~P1-1(dedup rc=1 静默转成功)~~ 已修复(见上方小节);余:守护进程补跑 steps 含 cninfo_*、调度时区与 date.today() 一致性,待用户决策
 - 既有小问题:`tests/test_scheduler.py` 部分 run_pipeline 测试未传 state_path,会写真实 `data/pipeline/state.json`(既有行为,未改)
 
 ---
diff --git a/scheduler/pipeline.py b/scheduler/pipeline.py
index ec8c7e0..a025162 100644
--- a/scheduler/pipeline.py
+++ b/scheduler/pipeline.py
@@ -290,11 +290,6 @@ def run_step(name: str, date_str: str) -> StepResult:
         elapsed = (datetime.now() - started).total_seconds()
         ok = proc.returncode == 0
 
-        # dedup 返回 1 是"重复率过高"(无新文章的正常场景)
-        if name == "dedup" and proc.returncode == 1:
-            ok = True
-            logger.info("dedup 重复率超过阈值(无新数据场景,视为成功)")
-
         tail_msg = f"rc={proc.returncode}" if not ok else ""
 
         if ok:
diff --git a/scripts/run_dedup.py b/scripts/run_dedup.py
index c84e93d..ec9b9d2 100644
--- a/scripts/run_dedup.py
+++ b/scripts/run_dedup.py
@@ -22,7 +22,7 @@ import argparse
 import json
 import sys
 from collections import Counter
-from datetime import date
+from datetime import date, datetime
 from pathlib import Path
 
 from loguru import logger
@@ -31,6 +31,9 @@ from pydantic import ValidationError
 from dedup import Deduper
 from extractor import Article
 
+# 重复率告警阈值(超过则输出 WARNING 日志;--strict 验收模式下作为返回码门槛)
+_DUP_RATE_THRESHOLD = 0.05
+
 
 def _setup_logger(level: str) -> None:
     logger.remove()
@@ -200,6 +203,11 @@ def main() -> int:
     parser.add_argument("--simhash-threshold", type=int, default=3)
     parser.add_argument("--window-days", type=int, default=30)
     parser.add_argument("--reset", action="store_true", help="处理前清空指纹库")
+    parser.add_argument(
+        "--strict", action="store_true",
+        help="验收模式:重复率超过阈值(5%)时返回 1。"
+             "默认为生产模式,重复率仅作统计告警,不影响退出码",
+    )
     parser.add_argument("--log-level", default="INFO")
     args = parser.parse_args()
 
@@ -243,6 +251,7 @@ def main() -> int:
         # {url_hash: [source_id, ...]},配合 uniques/{url_hash}.json 的 sources 字段
         # 与指纹库 source_ids 列,提供「一条唯一新闻多个来源」的完整记录。
         sources_path = out_root / args.date / "sources.json"
+        sources_path.parent.mkdir(parents=True, exist_ok=True)
         sources_path.write_text(
             json.dumps(
                 {k: v for k, v in sources_map.items() if v},
@@ -260,15 +269,47 @@ def main() -> int:
 
         total = total_uniq + total_dup
         rate = total_dup / max(total, 1)
-        logger.info(
+        log_fn = logger.warning if rate > _DUP_RATE_THRESHOLD else logger.info
+        log_fn(
             "全部完成: 唯一 {} / 重复 {} (重复率 {:.1%}) layers={}",
             total_uniq,
             total_dup,
             rate,
             dict(total_layers),
         )
-        # 验收门槛: ≤ 5%
-        return 0 if rate <= 0.05 or total == 0 else 1
+
+        # 统计快照: data/deduped/{date}/stats.json(原子写),供运维排查与监控
+        stats_payload = {
+            "date": args.date,
+            "generated_at": datetime.now().isoformat(),
+            "unique": total_uniq,
+            "duplicates": total_dup,
+            "total": total,
+            "dup_rate": round(rate, 4),
+            "dup_rate_threshold": _DUP_RATE_THRESHOLD,
+            "layers": dict(total_layers),
+            "fingerprint_total": deduper.stats().total,
+        }
+        stats_path = out_root / args.date / "stats.json"
+        stats_path.parent.mkdir(parents=True, exist_ok=True)
+        tmp = stats_path.with_suffix(".json.tmp")
+        tmp.write_text(json.dumps(stats_payload, ensure_ascii=False, indent=2),
+                       encoding="utf-8")
+        tmp.replace(stats_path)
+
+        if rate > _DUP_RATE_THRESHOLD and total > 0:
+            logger.warning(
+                "重复率 {:.1%} 超过阈值 {:.0%}(唯一 {},重复 {})。"
+                "同日多次调度/跨源转载属常见现象;若为当日首次处理仍异常偏高,"
+                "请检查指纹库与抓取源(详见 {})",
+                rate, _DUP_RATE_THRESHOLD, total_uniq, total_dup, stats_path,
+            )
+
+        # 生产模式:重复率是统计指标,不影响退出码(执行成功即 0);
+        # 验收模式(--strict):保留 M3 验收门槛(重复率 ≤ 5%),超标返回 1。
+        if args.strict and total > 0 and rate > _DUP_RATE_THRESHOLD:
+            return 1
+        return 0
 
 
 if __name__ == "__main__":
diff --git a/tests/test_run_dedup.py b/tests/test_run_dedup.py
new file mode 100644
index 0000000..24ecc47
--- /dev/null
+++ b/tests/test_run_dedup.py
@@ -0,0 +1,173 @@
+"""run_dedup 脚本测试(生产/验收返回码语义 + stats.json 统计快照)。
+
+覆盖:
+    - 生产模式(默认):高重复率不影响退出码,返回 0;重复率仅作告警
+    - 验收模式(--strict):高重复率返回 1(保留 M3 验收门槛 ≤ 5%)
+    - 低重复率场景两种模式均返回 0
+    - stats.json 统计快照结构正确(原子写),含重复率与指纹库总量
+"""
+
+from __future__ import annotations
+
+import hashlib
+import json
+import sys
+from pathlib import Path
+
+import pytest
+
+
+def _url_hash(url: str) -> str:
+    return hashlib.sha1(url.encode("utf-8")).hexdigest()[:16]
+
+
+def _write_article(proc_dir: Path, url: str, title: str, content: str) -> None:
+    """写一个 Article JSON,文件名为 url_hash。"""
+    h = _url_hash(url)
+    article = {
+        "source_id": "testsrc",
+        "url": url,
+        "url_hash": h,
+        "title": title,
+        "content": content,
+        "word_count": len(content),
+    }
+    (proc_dir / f"{h}.json").write_text(
+        json.dumps(article, ensure_ascii=False), encoding="utf-8"
+    )
+
+
+def _setup_fixture(tmp_path: Path, *, dup_count: int, uniq_count: int) -> Path:
+    """构造 processed 夹具:uniq_count 篇互不相同 + dup_count 篇与第 1 篇内容相同。"""
+    day = "20260616"
+    proc_dir = tmp_path / "processed" / "testsrc" / day
+    proc_dir.mkdir(parents=True)
+
+    base_content = "这是一条足够长的测试新闻正文内容,用于触发去重判定逻辑。" * 3
+    # 有重复篇时才写基准篇(重复目标);否则唯一数 = uniq_count
+    if dup_count > 0:
+        _write_article(proc_dir, "https://x.example/base", "基准新闻", base_content)
+        # 与基准篇内容相同、URL 不同 → 判为重复
+        for i in range(dup_count):
+            _write_article(
+                proc_dir, f"https://x.example/dup{i}", f"重复新闻{i}", base_content
+            )
+    # 互不相同的其它唯一文章
+    for i in range(uniq_count):
+        _write_article(
+            proc_dir, f"https://x.example/uniq{i}", f"独立新闻{i}",
+            f"完全不同的正文内容片段编号 {i},讲述另一件事。" * 3,
+        )
+    return proc_dir
+
+
+def _run_dedup_main(tmp_path: Path, *extra_args: str) -> int:
+    """以指定参数调用 scripts.run_dedup.main,返回退出码。"""
+    import scripts.run_dedup as mod
+
+    day = "20260616"
+    argv = [
+        "run_dedup",
+        "--processed-root", str(tmp_path / "processed"),
+        "--out-root", str(tmp_path / "deduped"),
+        "--db", str(tmp_path / "fingerprints.sqlite3"),
+        "--date", day,
+        "--log-level", "ERROR",
+        *extra_args,
+    ]
+    old_argv = sys.argv
+    sys.argv = argv
+    try:
+        return mod.main()
+    finally:
+        sys.argv = old_argv
+
+
+def _read_stats(tmp_path: Path) -> dict:
+    p = tmp_path / "deduped" / "20260616" / "stats.json"
+    assert p.is_file(), "stats.json 未生成"
+    return json.loads(p.read_text(encoding="utf-8"))
+
+
+# --------------------------------------------------------------------------- #
+# 返回码语义
+# --------------------------------------------------------------------------- #
+
+def test_production_mode_high_dup_returns_0(tmp_path: Path) -> None:
+    """生产模式:高重复率(90%)不影响退出码,返回 0。"""
+    _setup_fixture(tmp_path, dup_count=9, uniq_count=0)
+    rc = _run_dedup_main(tmp_path)
+    assert rc == 0, "生产模式下重复率仅作告警,不应改变退出码"
+
+
+def test_strict_mode_high_dup_returns_1(tmp_path: Path) -> None:
+    """验收模式 --strict:重复率超过 5% 门槛时返回 1。"""
+    _setup_fixture(tmp_path, dup_count=9, uniq_count=0)
+    rc = _run_dedup_main(tmp_path, "--strict")
+    assert rc == 1
+
+
+def test_strict_mode_low_dup_returns_0(tmp_path: Path) -> None:
+    """验收模式:重复率 ≤ 5% 时返回 0。"""
+    _setup_fixture(tmp_path, dup_count=0, uniq_count=10)
+    rc = _run_dedup_main(tmp_path, "--strict")
+    assert rc == 0
+
+
+def test_empty_day_returns_0(tmp_path: Path) -> None:
+    """当日无文章时返回 0(两种模式)。"""
+    proc_dir = tmp_path / "processed" / "testsrc" / "20260616"
+    proc_dir.mkdir(parents=True)
+    assert _run_dedup_main(tmp_path) == 0
+    assert _run_dedup_main(tmp_path, "--strict") == 0
+
+
+# --------------------------------------------------------------------------- #
+# stats.json 统计快照
+# --------------------------------------------------------------------------- #
+
+def test_stats_json_written_and_correct(tmp_path: Path) -> None:
+    """stats.json 结构正确:重复率、唯一/重复数、指纹库总量。"""
+    _setup_fixture(tmp_path, dup_count=9, uniq_count=0)
+    _run_dedup_main(tmp_path)
+    s = _read_stats(tmp_path)
+    assert s["date"] == "20260616"
+    assert s["unique"] == 1
+    assert s["duplicates"] == 9
+    assert s["total"] == 10
+    assert s["dup_rate"] == pytest.approx(0.9, abs=1e-6)
+    assert s["dup_rate_threshold"] == pytest.approx(0.05)
+    # 指纹库只保留唯一文章
+    assert s["fingerprint_total"] == 1
+    assert "generated_at" in s and "layers" in s
+
+
+def test_stats_json_low_dup_rate(tmp_path: Path) -> None:
+    """低重复率场景:dup_rate 为 0。"""
+    _setup_fixture(tmp_path, dup_count=0, uniq_count=5)
+    _run_dedup_main(tmp_path)
+    s = _read_stats(tmp_path)
+    assert s["unique"] == 5
+    assert s["duplicates"] == 0
+    assert s["dup_rate"] == 0.0
+    assert s["fingerprint_total"] == 5
+
+
+# --------------------------------------------------------------------------- #
+# pipeline: dedup 不再特判(真实失败现在会正确上报)
+# --------------------------------------------------------------------------- #
+
+def test_pipeline_dedup_failure_no_longer_masked(monkeypatch: pytest.MonkeyPatch) -> None:
+    """移除特判后:dedup 非 0 退出码应如实标记为失败。"""
+    from datetime import date
+    from types import SimpleNamespace
+
+    from scheduler import pipeline
+
+    def _fake_run(cmd, timeout=None):  # noqa: ARG001
+        return SimpleNamespace(returncode=1)
+
+    monkeypatch.setattr(pipeline.subprocess, "run", _fake_run)
+    sr = pipeline.run_step("dedup", date.today().strftime("%Y%m%d"))
+    assert sr.success is False, "dedup 返回非 0 应标记失败(特判已移除)"
+    assert "rc=1" in sr.tail_msg