update：更新skill脚本

config/config.example.yaml

@@ -6,3 +6,6 @@ feishu:
 hisense_geo:
   base_url: "https://geo-api.hisense.com"
   token: ""       # 生产环境 x-token，勿提交仓库
+haoxian:
+  base_url: "https://resource-api.cemeta.cn"
+  token: ""       # 好现 x-token，勿提交仓库

haoxian/skills/temp0923/SKILL.md

@@ -0,0 +1,119 @@
+---
+name: hisense-sept-qa-raw-query
+description: >-
+  Query Hisense September agentic projects for qa_raw and qa_product_raw
+  (HISENSE-HIST). Before any API call, must confirm BOTH batch date and project
+  id together. Returns id, project_name, total, qa_raw_list, qa_product_raw_list.
+  Use when user asks for 9月项目、问答底表、商品卡、qa_raw、qa_product_raw、
+  HISENSE-HIST、temp0923, or 批次日期 / 项目id.
+---
+
+# 海信 9 月 qa_raw / qa_product_raw 查询
+
+从 `https://resource-api.cemeta.cn` 拉 9 月项目，再按批次查问答底表与商品卡表，组装返回 JSON。
+
+配置：`config/config.dev.yaml` → `haoxian.base_url` / `haoxian.token`（Header `x-token`）。本机可用 `http://localhost:9999` 代替 base（路径相同）。
+
+## 必选前置输入（缺一不可，须同时确认）
+
+**使用本 skill 时，执行任何接口调用之前，必须先同时拿到下面 1 和 2。缺任意一项都停下来，两项齐了再继续后续流程。不要猜、不要沿用上次对话默认值、不要只拿到一个就先跑一半。**
+
+| # | 必选项 | 示例 |
+|---|--------|------|
+| 1 | 查询的批次日期 | `20260919`（`YYYYMMDD`） |
+| 2 | 项目 id | `1096` |
+
+**一次性原文提示用户（同一条消息里同时确认两项）：**
+
+> 请同时确认后回复：
+> 1. 查询的批次日期（示例：20260919）
+> 2. 项目id（示例：1096）
+
+用户在同一轮或连续回复中给出 1 和 2 后，再执行下方「执行流程」。若只给了一项，用同样提示补要缺失项，仍不要开查。
+
+## 执行流程
+
+### 1. 拉 9 月项目列表
+
+```
+GET {base_url}/api/agentic-project/project?page=1&page_size=999&listed=all
+Header: x-token
+```
+
+过滤（必须同时满足）：
+
+- `listed = 1`
+- `status = 'running'`
+- `end_month = '2026-09'`
+
+取出全部 `id`、`project_name`。用用户输入的项目 id 在列表中解析 `project_name`（找不到也继续查表，name 可为空并注明）。
+
+### 2. 拼接 crawl_batch_id
+
+```
+HISENSE-HIST-{project_id}-{批次日期}
+```
+
+例：`HISENSE-HIST-1096-20260919`
+
+### 3. 查问答底表 qa_raw
+
+```
+GET {base_url}/api/geo_label/l1/qa_raw?project_id={id}&crawl_batch_id={crawl_batch_id}&page=1&page_size=50
+Header: x-token
+```
+
+翻页直到收齐 `data.total`。接口列表字段为 `data.items`。
+
+### 4. 查商品卡表 qa_product_raw
+
+```
+GET {base_url}/api/geo_label/l1/qa_product_raw?project_id={id}&crawl_batch_id={crawl_batch_id}&page=1&page_size=50
+Header: x-token
+```
+
+同样翻页收齐；接口列表字段为 `data.items`。
+
+### 5. 组装返回 JSON
+
+**对外不要输出 `items`。** `data.items` 分别映射为：
+
+- qa_raw → `qa_raw_list`
+- qa_product_raw → `qa_product_raw_list`
+
+`total` 取 **qa_raw** 的 `data.total`。
+
+```json
+{
+  "id": 1096,
+  "project_name": "500L左右冰箱推荐",
+  "total": 227,
+  "code": 0,
+  "ok": true,
+  "crawl_batch_id": "HISENSE-HIST-1096-20260919",
+  "qa_raw_list": [],
+  "qa_product_raw_list": []
+}
+```
+
+必须包含：`id`、`project_name`、`total`、`qa_raw_list`（全量明细）、`qa_product_raw_list`（全量明细）。体积大时落盘并告知路径，例如：
+
+`haoxian/skills/temp0923/data/qa_raw_{批次}/{id}_response.json`
+
+## 推荐脚本
+
+```bash
+# 可选：刷新 9 月项目缓存
+python3 haoxian/skills/temp0923/scripts/query_qa_raw.py list-projects
+
+# 必选参数：--batch、--id
+python3 haoxian/skills/temp0923/scripts/query_qa_raw.py project --batch 20260919 --id 1096
+```
+
+脚本 stdout 即为上述返回结构。本地缓存目录：`haoxian/skills/temp0923/data/`。
+
+## 注意
+
+- Token 只读配置，勿写入 skill、勿提交仓库。
+- `total=0` 或空列表均为有效结果。
+- 翻页建议 `page_size=50`，失败重试，避免一次过大导致断连。

haoxian/skills/temp0923/scripts/query_qa_raw.py

@@ -0,0 +1,348 @@
+#!/usr/bin/env python3
+"""海信 9 月项目 + qa_raw / qa_product_raw 查询（resource-api / localhost）。"""
+
+from __future__ import annotations
+
+import argparse
+import json
+import sys
+import time
+import urllib.request
+from pathlib import Path
+
+ROOT = Path(__file__).resolve().parents[4]  # ai_script
+SKILL_DATA = Path(__file__).resolve().parents[1] / "data"
+DEFAULT_BASE = "https://resource-api.cemeta.cn"
+REQUIRED_END_MONTH = "2026-09"
+
+
+def load_haoxian_config() -> dict:
+    cfg_path = ROOT / "config" / "config.dev.yaml"
+    if not cfg_path.exists():
+        raise SystemExit(f"缺少配置: {cfg_path}")
+    section = None
+    hao: dict[str, str] = {}
+    for line in cfg_path.read_text().splitlines():
+        if line.startswith("haoxian:"):
+            section = "hao"
+            continue
+        if line and not line.startswith(" ") and line.endswith(":"):
+            section = None
+            continue
+        if section == "hao" and ":" in line and line.startswith("  "):
+            k, v = line.split(":", 1)
+            hao[k.strip()] = v.strip().strip('"').strip("'")
+    if not hao.get("token"):
+        raise SystemExit("config.dev.yaml 缺少 haoxian.token")
+    if not hao.get("base_url"):
+        hao["base_url"] = DEFAULT_BASE
+    return hao
+
+
+def http_get(url: str, token: str, retries: int = 4) -> dict:
+    last: Exception | None = None
+    for i in range(retries):
+        try:
+            req = urllib.request.Request(url, headers={"x-token": token})
+            with urllib.request.urlopen(req, timeout=180) as resp:
+                return json.loads(resp.read())
+        except Exception as e:  # noqa: BLE001
+            last = e
+            time.sleep(1.5 * (i + 1))
+    raise RuntimeError(f"GET failed after retries: {url}: {last}")
+
+
+def list_sept_projects(base: str, token: str) -> list[dict]:
+    url = f"{base.rstrip('/')}/api/agentic-project/project?page=1&page_size=999&listed=all"
+    data = http_get(url, token)
+    items = (data.get("data") or {}).get("list") or []
+    out = []
+    for p in items:
+        if (
+            p.get("listed") == 1
+            and str(p.get("status") or "").lower() == "running"
+            and p.get("end_month") == REQUIRED_END_MONTH
+        ):
+            out.append(
+                {
+                    "id": p.get("id"),
+                    "project_name": p.get("project_name"),
+                    "brand": p.get("brand"),
+                    "category": p.get("category"),
+                    "end_month": p.get("end_month"),
+                    "status": p.get("status"),
+                    "listed": p.get("listed"),
+                }
+            )
+    out.sort(key=lambda x: int(x["id"]) if str(x["id"]).isdigit() else 0, reverse=True)
+    return out
+
+
+def fetch_l1_table(
+    base: str,
+    token: str,
+    table: str,
+    project_id: int | str,
+    batch: str,
+    page_size: int = 50,
+) -> dict:
+    """分页拉取 /api/geo_label/l1/{table}，返回 {total, code, ok, crawl_batch_id, items}。"""
+    crawl = f"HISENSE-HIST-{project_id}-{batch}"
+    all_items: list = []
+    total = None
+    code = 0
+    page = 1
+    while True:
+        url = (
+            f"{base.rstrip('/')}/api/geo_label/l1/{table}"
+            f"?project_id={project_id}&crawl_batch_id={crawl}"
+            f"&page={page}&page_size={page_size}"
+        )
+        data = http_get(url, token)
+        code = data.get("code", 0)
+        if code != 0:
+            return {
+                "id": int(project_id) if str(project_id).isdigit() else project_id,
+                "total": 0,
+                "code": code,
+                "ok": False,
+                "crawl_batch_id": crawl,
+                "items": [],
+                "message": data.get("message"),
+            }
+        d = data.get("data") or {}
+        if total is None:
+            total = d.get("total", 0) or 0
+        page_items = d.get("items") or []
+        all_items.extend(page_items)
+        if len(all_items) >= total or not page_items:
+            break
+        page += 1
+        if page > 100:
+            break
+    return {
+        "id": int(project_id) if str(project_id).isdigit() else project_id,
+        "total": total,
+        "code": code,
+        "ok": True,
+        "crawl_batch_id": crawl,
+        "fetched": len(all_items),
+        "items": all_items,
+    }
+
+
+def fetch_qa_raw(base: str, token: str, project_id: int | str, batch: str, page_size: int = 50) -> dict:
+    return fetch_l1_table(base, token, "qa_raw", project_id, batch, page_size)
+
+
+def fetch_qa_product_raw(base: str, token: str, project_id: int | str, batch: str, page_size: int = 50) -> dict:
+    return fetch_l1_table(base, token, "qa_product_raw", project_id, batch, page_size)
+
+
+def cmd_list_projects(args: argparse.Namespace) -> None:
+    cfg = load_haoxian_config()
+    base = args.base or cfg["base_url"]
+    projects = list_sept_projects(base, cfg["token"])
+    SKILL_DATA.mkdir(parents=True, exist_ok=True)
+    out = SKILL_DATA / "sept_projects_202609.json"
+    out.write_text(json.dumps(projects, ensure_ascii=False, indent=2))
+    print(json.dumps({"count": len(projects), "path": str(out), "projects": projects}, ensure_ascii=False, indent=2))
+
+
+def cmd_batch_totals(args: argparse.Namespace) -> None:
+    cfg = load_haoxian_config()
+    base = args.base or cfg["base_url"]
+    batch = args.batch
+    projects = list_sept_projects(base, cfg["token"])
+    cache_dir = SKILL_DATA / f"qa_raw_{batch}"
+    cache_dir.mkdir(parents=True, exist_ok=True)
+
+    rows = []
+    for i, p in enumerate(projects, 1):
+        pid = p["id"]
+        cache = cache_dir / f"{pid}.json"
+        if cache.exists() and not args.refresh:
+            payload = json.loads(cache.read_text())
+            total = payload.get("total", 0)
+            items = payload.get("items") or []
+            row = {
+                "id": pid,
+                "project_name": p["project_name"],
+                "total": total,
+                "code": 0,
+                "ok": True,
+                "crawl_batch_id": payload.get("crawl_batch_id") or f"HISENSE-HIST-{pid}-{batch}",
+                "fetched": len(items),
+            }
+        else:
+            try:
+                raw = fetch_qa_raw(base, cfg["token"], pid, batch, page_size=args.page_size)
+            except Exception as e:  # noqa: BLE001
+                row = {
+                    "id": pid,
+                    "project_name": p["project_name"],
+                    "total": -1,
+                    "code": None,
+                    "ok": False,
+                    "err": str(e),
+                }
+                rows.append(row)
+                print(f"[{i}/{len(projects)}] FAIL id={pid}: {e}", file=sys.stderr)
+                continue
+            items = raw.get("items") or []
+            payload = {
+                "project_id": pid,
+                "project_name": p["project_name"],
+                "crawl_batch_id": raw["crawl_batch_id"],
+                "batch_date": batch,
+                "total": raw["total"],
+                "fetched": raw.get("fetched", len(items)),
+                "items": items,
+            }
+            cache.write_text(json.dumps(payload, ensure_ascii=False))
+            row = {
+                "id": pid,
+                "project_name": p["project_name"],
+                "total": raw["total"],
+                "code": raw.get("code", 0),
+                "ok": raw.get("ok", True),
+                "crawl_batch_id": raw["crawl_batch_id"],
+                "fetched": payload["fetched"],
+            }
+        rows.append(row)
+        if i % 20 == 0 or i == len(projects):
+            print(f"[{i}/{len(projects)}] ...", file=sys.stderr)
+
+    summary = {
+        "batch_date": batch,
+        "base_url": base,
+        "count": len(rows),
+        "ok": sum(1 for r in rows if r.get("ok")),
+        "total_sum": sum(r["total"] for r in rows if r.get("ok") and r["total"] >= 0),
+        "projects": rows,
+    }
+    (SKILL_DATA / f"qa_raw_{batch}_summary.json").write_text(json.dumps(summary, ensure_ascii=False, indent=2))
+    tsv = ["project_id\tproject_name\ttotal\tok\tcrawl_batch_id"]
+    for r in rows:
+        tsv.append(
+            f"{r.get('id')}\t{r.get('project_name')}\t{r.get('total')}\t{r.get('ok')}\t{r.get('crawl_batch_id','')}"
+        )
+    (SKILL_DATA / f"qa_raw_{batch}_totals.tsv").write_text("\n".join(tsv) + "\n")
+    print(json.dumps({"batch_date": batch, "count": len(rows), "projects": rows}, ensure_ascii=False, indent=2))
+
+
+def cmd_project(args: argparse.Namespace) -> None:
+    cfg = load_haoxian_config()
+    base = args.base or cfg["base_url"]
+    batch = args.batch
+    pid = args.id
+    cache_dir = SKILL_DATA / f"qa_raw_{batch}"
+    cache = cache_dir / f"{pid}.json"
+    product_cache = cache_dir / f"{pid}_product_raw.json"
+
+    name = None
+    projects_path = SKILL_DATA / "sept_projects_202609.json"
+    if projects_path.exists():
+        for p in json.loads(projects_path.read_text()):
+            if str(p.get("id")) == str(pid):
+                name = p.get("project_name")
+                break
+
+    if cache.exists() and not args.refresh:
+        payload = json.loads(cache.read_text())
+        qa_raw_list = payload.get("items") or payload.get("qa_raw_list") or []
+        crawl = payload.get("crawl_batch_id") or f"HISENSE-HIST-{pid}-{batch}"
+        total = payload.get("total")
+        code = 0
+        ok = True
+        if not name:
+            name = payload.get("project_name")
+    else:
+        raw = fetch_qa_raw(base, cfg["token"], pid, batch, page_size=args.page_size)
+        qa_raw_list = raw.get("items") or []
+        crawl = raw["crawl_batch_id"]
+        total = raw["total"]
+        code = raw.get("code", 0)
+        ok = raw.get("ok", True)
+        cache_dir.mkdir(parents=True, exist_ok=True)
+        cache.write_text(
+            json.dumps(
+                {
+                    "project_id": raw["id"],
+                    "project_name": name,
+                    "crawl_batch_id": crawl,
+                    "batch_date": batch,
+                    "total": total,
+                    "fetched": len(qa_raw_list),
+                    "items": qa_raw_list,
+                },
+                ensure_ascii=False,
+            )
+        )
+
+    if product_cache.exists() and not args.refresh:
+        prod_payload = json.loads(product_cache.read_text())
+        qa_product_raw_list = (
+            prod_payload.get("items") or prod_payload.get("qa_product_raw_list") or []
+        )
+    else:
+        prod = fetch_qa_product_raw(base, cfg["token"], pid, batch, page_size=args.page_size)
+        qa_product_raw_list = prod.get("items") or []
+        if not prod.get("ok", True):
+            ok = False
+            code = prod.get("code", code)
+        cache_dir.mkdir(parents=True, exist_ok=True)
+        product_cache.write_text(
+            json.dumps(
+                {
+                    "project_id": int(pid) if str(pid).isdigit() else pid,
+                    "project_name": name,
+                    "crawl_batch_id": crawl,
+                    "batch_date": batch,
+                    "total": prod.get("total", len(qa_product_raw_list)),
+                    "fetched": len(qa_product_raw_list),
+                    "items": qa_product_raw_list,
+                },
+                ensure_ascii=False,
+            )
+        )
+
+    out = {
+        "id": int(pid) if str(pid).isdigit() else pid,
+        "project_name": name,
+        "total": total,
+        "code": code,
+        "ok": ok,
+        "crawl_batch_id": crawl,
+        "qa_raw_list": qa_raw_list,
+        "qa_product_raw_list": qa_product_raw_list,
+    }
+    print(json.dumps(out, ensure_ascii=False, indent=2))
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="Hisense Sept qa_raw query")
+    parser.add_argument("--base", default=None, help="API base, default from config / resource-api")
+    parser.add_argument("--page-size", type=int, default=50)
+    sub = parser.add_subparsers(dest="cmd", required=True)
+
+    p1 = sub.add_parser("list-projects", help="List Sept running projects")
+    p1.set_defaults(func=cmd_list_projects)
+
+    p2 = sub.add_parser("batch-totals", help="Fetch/totals for all Sept projects in a batch")
+    p2.add_argument("--batch", required=True, help="Batch date YYYYMMDD, e.g. 20260919")
+    p2.add_argument("--refresh", action="store_true")
+    p2.set_defaults(func=cmd_batch_totals)
+
+    p3 = sub.add_parser("project", help="Fetch qa_raw + qa_product_raw; stdout uses qa_raw_list")
+    p3.add_argument("--batch", required=True)
+    p3.add_argument("--id", required=True)
+    p3.add_argument("--refresh", action="store_true")
+    p3.set_defaults(func=cmd_project)
+
+    args = parser.parse_args()
+    args.func(args)
+
+
+if __name__ == "__main__":
+    main()