diff --git a/backend/crawler_copy/jd_pc_search/jd_competitor_report.py b/backend/crawler_copy/jd_pc_search/jd_competitor_report.py
index c6b6d6f..f457981 100644
--- a/backend/crawler_copy/jd_pc_search/jd_competitor_report.py
+++ b/backend/crawler_copy/jd_pc_search/jd_competitor_report.py
@@ -1,9 +1,11 @@
 # -*- coding: utf-8 -*-
 """
-兼容入口（历史路径）：竞品报告实现已迁至 ``pipeline.jd_competitor_report``。
+在爬虫目录下直接执行时的入口：将 ``backend`` 加入 ``sys.path`` 后调用
+``pipeline.competitor_report.jd_report``（与 ``python -m pipeline.competitor_report.jd_report`` 一致）。
 
-- 推荐：在 ``backend`` 目录执行 ``python -m pipeline.jd_competitor_report``
-- 本文件：将 ``backend`` 加入 ``sys.path`` 后转发至 ``pipeline`` 模块，便于仍在爬虫目录下执行 ``python jd_competitor_report.py`` 的旧习惯。
+推荐在 ``backend`` 目录执行::
+
+  python -m pipeline.competitor_report.jd_report
 """
 from __future__ import annotations
 
@@ -14,4 +16,7 @@ _backend = Path(__file__).resolve().parents[2]
 if str(_backend) not in sys.path:
     sys.path.insert(0, str(_backend))
 
-from pipeline.jd_competitor_report import *  # noqa: F403
+from pipeline.competitor_report import jd_report as _jd  # noqa: E402
+
+if __name__ == "__main__":
+    _jd.main()
diff --git a/backend/crawler_copy/jd_pc_search/jd_keyword_pipeline.py b/backend/crawler_copy/jd_pc_search/jd_keyword_pipeline.py
index e83de89..090068e 100644
--- a/backend/crawler_copy/jd_pc_search/jd_keyword_pipeline.py
+++ b/backend/crawler_copy/jd_pc_search/jd_keyword_pipeline.py
@@ -25,7 +25,7 @@ PC 搜索导出 CSV、评价扁平 CSV、详情汇总 CSV（``detail_ware_export
 默认启用 **应用场景筛选**（``brief_content.txt`` 4.1 中式面点/主食 + 4.2 烘焙）：仅命中关键词的 SKU 进入详情与评论队列；词表见 ``scenario_filter.py``。``SCENARIO_FILTER_ENABLED=False`` 可关闭；``SCENARIO_FILTER_PC_SEARCH_CSV="filtered"`` 可使导出 CSV 与筛选后列表一致。
 各 SKU 完整接口 JSON 仍在 ``detail/ware_{sku}_response.json``。
 
-端到端竞品速览 Markdown：在 ``backend`` 下配置 ``pipeline.jd_competitor_report`` 顶部 ``KEYWORD`` 后执行 ``python -m pipeline.jd_competitor_report``；或执行本目录兼容入口 ``python jd_competitor_report.py``（内部调用本模块 ``main(keyword=...)``）。
+端到端竞品速览 Markdown：在 ``backend`` 下配置 ``pipeline.competitor_report.jd_report`` 顶部 ``KEYWORD`` 后执行 ``python -m pipeline.competitor_report.jd_report``；或执行本目录兼容入口 ``python jd_competitor_report.py``（调用同一 ``jd_report.main``）。
 """
 
 from __future__ import annotations
@@ -246,7 +246,7 @@ def main(keyword: str | None = None) -> Path:
     """
     跑完整条流水线。``keyword`` 非空时覆盖文件内 ``KEYWORD``；返回本次运行目录。
 
-    供 ``pipeline.jd_competitor_report`` 等脚本 ``import`` 调用；命令行仍执行 ``main()`` 无参。
+    供 ``pipeline.competitor_report.jd_report`` 等脚本 ``import`` 调用；命令行仍执行 ``main()`` 无参。
     """
     try:
         if hasattr(sys.stdout, "reconfigure"):
diff --git a/backend/pipeline/competitor_report/constants.py b/backend/pipeline/competitor_report/constants.py
index 7b2c199..821d433 100644
--- a/backend/pipeline/competitor_report/constants.py
+++ b/backend/pipeline/competitor_report/constants.py
@@ -1,4 +1,4 @@
-"""CSV 表头键、运行默认调参与关注词/场景配置（与 ``pipeline.jd_competitor_report`` 顶层一致）。"""
+"""CSV 表头键、运行默认调参与关注词/场景配置（与 ``pipeline.competitor_report.jd_report`` 顶层一致）。"""
 from __future__ import annotations
 
 from pipeline.csv_schema import (
diff --git a/backend/pipeline/competitor_report/jd_report.py b/backend/pipeline/competitor_report/jd_report.py
index b0c6a4c..3a6d746 100644
--- a/backend/pipeline/competitor_report/jd_report.py
+++ b/backend/pipeline/competitor_report/jd_report.py
@@ -10,12 +10,11 @@
 **仅复用已有目录生成报告时**不需要跑浏览器，只需该目录下已有 CSV / ``run_meta.json``。
 
 实现位于 ``pipeline.competitor_report.jd_report``；爬虫仅在 ``crawler_copy/jd_pc_search``。
-根目录 ``pipeline.jd_competitor_report`` 为兼容入口。
 
 用法：
 
 - **重新抓取并出报告**：``EXISTING_RUN_DIR = None``，配置 ``KEYWORD``（及可选 ``OVERRIDE_*``），在 ``backend`` 目录下执行
-  ``python -m pipeline.jd_competitor_report``；或沿用爬虫目录下的兼容入口 ``python jd_competitor_report.py``（见该文件说明）。
+  ``python -m pipeline.competitor_report.jd_report``；或沿用爬虫目录下的兼容入口 ``python jd_competitor_report.py``（见该文件说明）。
 - **只分析已有批次**：将 ``EXISTING_RUN_DIR`` 设为 ``pipeline_runs/<时间戳>_<关键词>/`` 的绝对或相对路径（相对当前工作目录），
   再执行同一命令；**不重新抓取**。关键词优先用本文件 ``KEYWORD``，否则读 ``run_meta.json`` 的 ``keyword``，再否则从目录名
   ``YYYYMMDD_HHMMSS_<词>`` 推断。
diff --git a/backend/pipeline/demos/chapter8_text_mining_probe.py b/backend/pipeline/demos/chapter8_text_mining_probe.py
index 0212b48..a79dffe 100644
--- a/backend/pipeline/demos/chapter8_text_mining_probe.py
+++ b/backend/pipeline/demos/chapter8_text_mining_probe.py
@@ -43,7 +43,7 @@ JCR_ROOT = BACKEND_ROOT / "crawler_copy" / "jd_pc_search"
 if str(JCR_ROOT) not in sys.path:
     sys.path.insert(0, str(JCR_ROOT))
 
-from pipeline import jd_competitor_report as jcr  # noqa: E402
+from pipeline.competitor_report import jd_report as jcr  # noqa: E402
 import jd_keyword_pipeline as kpl  # noqa: E402
 
 from pipeline.csv_schema import MERGED_FIELD_TO_CSV_HEADER  # noqa: E402
diff --git a/backend/pipeline/demos/run_price_groups_llm_demo.py b/backend/pipeline/demos/run_price_groups_llm_demo.py
index 6c52244..ba9a213 100644
--- a/backend/pipeline/demos/run_price_groups_llm_demo.py
+++ b/backend/pipeline/demos/run_price_groups_llm_demo.py
@@ -26,7 +26,7 @@ JCR_ROOT = BACKEND_ROOT / "crawler_copy" / "jd_pc_search"
 if str(JCR_ROOT) not in sys.path:
     sys.path.insert(0, str(JCR_ROOT))
 
-from pipeline import jd_competitor_report as jcr  # noqa: E402
+from pipeline.competitor_report import jd_report as jcr  # noqa: E402
 import jd_keyword_pipeline as kpl  # noqa: E402
 
 
diff --git a/backend/pipeline/demos/run_report_llm_chapters_demo.py b/backend/pipeline/demos/run_report_llm_chapters_demo.py
index e491d7e..b64a301 100644
--- a/backend/pipeline/demos/run_report_llm_chapters_demo.py
+++ b/backend/pipeline/demos/run_report_llm_chapters_demo.py
@@ -37,7 +37,7 @@ JCR_ROOT = BACKEND_ROOT / "crawler_copy" / "jd_pc_search"
 if str(JCR_ROOT) not in sys.path:
     sys.path.insert(0, str(JCR_ROOT))
 
-from pipeline import jd_competitor_report as jcr  # noqa: E402
+from pipeline.competitor_report import jd_report as jcr  # noqa: E402
 import jd_keyword_pipeline as kpl  # noqa: E402
 
 from pipeline.csv_schema import MERGED_FIELD_TO_CSV_HEADER  # noqa: E402
diff --git a/backend/pipeline/jd/runner.py b/backend/pipeline/jd/runner.py
index 0ad756d..b0599f0 100644
--- a/backend/pipeline/jd/runner.py
+++ b/backend/pipeline/jd/runner.py
@@ -1,5 +1,5 @@
 """
-使用 ``crawler_copy/jd_pc_search`` 中的采集脚本执行流水线；竞品 Markdown 由 ``pipeline.jd_competitor_report`` 生成。
+使用 ``crawler_copy/jd_pc_search`` 中的采集脚本执行流水线；竞品 Markdown 由 ``pipeline.competitor_report.jd_report`` 生成。
 依赖环境变量 ``LOW_GI_PROJECT_ROOT``（由 Django settings 从 ``market_assistant/.env`` 注入）。
 """
 from __future__ import annotations
@@ -133,7 +133,7 @@ def try_write_competitor_report_if_merged_exists(
 
 
 def _jd_crawler_modules():
-    from pipeline import jd_competitor_report as jcr  # noqa: WPS433
+    from pipeline.competitor_report import jd_report as jcr  # noqa: WPS433
 
     root = Path(settings.CRAWLER_JD_ROOT)
     if not root.is_dir():
@@ -164,7 +164,7 @@ def use_chunked_group_summaries_llm(report_config: dict[str, Any] | None) -> boo
 
 
 def get_default_report_config() -> dict[str, Any]:
-    """与 ``pipeline.jd_competitor_report`` 模块常量一致的默认报告调参（供前端回填）。"""
+    """与 ``pipeline.competitor_report.jd_report`` 模块常量一致的默认报告调参（供前端回填）。"""
     jcr, _ = _jd_crawler_modules()
     return {
         "llm_comment_sentiment": True,
diff --git a/backend/pipeline/jd_competitor_report.py b/backend/pipeline/jd_competitor_report.py
deleted file mode 100644
index 13cee18..0000000
--- a/backend/pipeline/jd_competitor_report.py
+++ /dev/null
@@ -1,34 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-兼容入口：竞品报告实现见 ``pipeline.competitor_report.jd_report``。
-
-命令行：``python -m pipeline.jd_competitor_report``；爬虫目录下的 ``jd_competitor_report.py`` 仍转发至此。
-"""
-from __future__ import annotations
-
-import pipeline.competitor_report.jd_report as _jd
-
-# ``from jd_report import *`` 不会绑定以下划线开头的名称；外部常以
-# ``import pipeline.jd_competitor_report as jcr`` 再访问 ``jcr._read_csv_rows`` 等。
-# 这里把实现模块中除模块身份相关 dunder 以外的符号全部同步到本模块，避免再漏导出。
-_SKIP = frozenset(
-    {
-        "__name__",
-        "__doc__",
-        "__file__",
-        "__package__",
-        "__loader__",
-        "__spec__",
-        "__path__",
-        "__cached__",
-        "__builtins__",
-        "__annotations__",
-    }
-)
-for _k, _v in vars(_jd).items():
-    if _k in _SKIP or (_k.startswith("__") and _k.endswith("__")):
-        continue
-    globals()[_k] = _v
-
-if __name__ == "__main__":
-    _jd.main()
diff --git a/backend/pipeline/reporting/charts.py b/backend/pipeline/reporting/charts.py
index 0f61ed9..021760c 100644
--- a/backend/pipeline/reporting/charts.py
+++ b/backend/pipeline/reporting/charts.py
@@ -545,7 +545,7 @@ def generate_report_charts(
     )
 
     def scenario_group_asset_slug(group: str, index: int) -> str:
-        """与 ``pipeline.jd_competitor_report`` / ``report_md_helpers._scenario_group_asset_slug`` 保持一致。"""
+        """与 ``pipeline.competitor_report.jd_report`` / ``report_md_helpers._scenario_group_asset_slug`` 保持一致。"""
         raw = (group or "").strip()
         core = re.sub(r"[^\w\u4e00-\u9fff-]", "", raw)[:20]
         if not core:
diff --git a/backend/pipeline/tests/test_comment_group_summaries_llm.py b/backend/pipeline/tests/test_comment_group_summaries_llm.py
index d89bf7a..9a69dd0 100644
--- a/backend/pipeline/tests/test_comment_group_summaries_llm.py
+++ b/backend/pipeline/tests/test_comment_group_summaries_llm.py
@@ -13,7 +13,7 @@ from pipeline.llm.generate import (
 
 
 def _load_jcr():
-    from pipeline import jd_competitor_report as jcr  # noqa: WPS433
+    from pipeline.competitor_report import jd_report as jcr  # noqa: WPS433
 
     return jcr
 
diff --git a/backend/pipeline/tests/test_competitor_brief.py b/backend/pipeline/tests/test_competitor_brief.py
index 80b3221..79b416a 100644
--- a/backend/pipeline/tests/test_competitor_brief.py
+++ b/backend/pipeline/tests/test_competitor_brief.py
@@ -6,7 +6,7 @@ from pathlib import Path
 
 from django.test import SimpleTestCase
 
-from pipeline import jd_competitor_report as jcr
+from pipeline.competitor_report import jd_report as jcr
 from pipeline.csv_schema import infer_total_sales_from_sales_floor
 from pipeline.reporting.charts import _cn_volume_int
 
diff --git a/backend/pipeline/views/job_views.py b/backend/pipeline/views/job_views.py
index bf2cff8..7fb408b 100644
--- a/backend/pipeline/views/job_views.py
+++ b/backend/pipeline/views/job_views.py
@@ -195,7 +195,7 @@ class JobResumeView(APIView):
 
 
 class ReportConfigDefaultsView(APIView):
-    """返回 ``jd_competitor_report`` 中与脚本常量一致的默认报告调参 JSON。"""
+    """返回 ``pipeline.competitor_report.jd_report`` 中与脚本常量一致的默认报告调参 JSON。"""
 
     def get(self, request):
         try:
diff --git a/docs/openapi/pipeline-jobs.openapi.yaml b/docs/openapi/pipeline-jobs.openapi.yaml
index 89fd5f7..5c80754 100644
--- a/docs/openapi/pipeline-jobs.openapi.yaml
+++ b/docs/openapi/pipeline-jobs.openapi.yaml
@@ -112,7 +112,7 @@ paths:
   /api/report-config-defaults/:
     get:
       tags: [jobs]
-      summary: 报告调参默认模板（与 jd_competitor_report 脚本常量一致）
+      summary: 报告调参默认模板（与 pipeline.competitor_report.jd_report 脚本常量一致）
       responses:
         "200":
           description: OK
@@ -501,7 +501,7 @@ components:
           type: string
           enum: [rules, llm]
           default: rules
-          description: rules=jd_competitor_report 规则引擎；llm=结构化摘要 JSON + AI_crawler 文本接口
+          description: rules=竞品报告规则引擎（jd_report）；llm=结构化摘要 JSON + AI_crawler 文本接口
 
     StrategyDraftRequest:
       type: object