feat: alert 补全 flow 名;AGENTS.md 补充 monitor 架构说明

AGENTS.md

@@ -79,6 +79,15 @@ uv run prefect <cmd>         # 运行 prefect CLI
 
 > 注:`serve_all.py` 为旧进程式部署,已被 work pool 模型取代,保留仅作参考。
 
+## 企微机器人告警/上报(monitor)
+
+- 发送层 `flows/shared/wecom.py`:`send_markdown/send_text`,wecom-aibot-sdk 按需连接(连接→认证→发送→断开),全容错。
+  凭据在 `Secret` Block `wecom-aibot-credentials`(JSON:`bot_id/secret/chat_id`);chat_id 用 `scripts/wecom_capture_chatid.py` 捕获。
+- `flows/monitor/alert_failure.py`(wecom-alert):prefect.yaml 里声明的事件触发器,任意 flow run `Failed/Crashed` 自动运行并发群报警,零代码侵入,新 flow 自动覆盖。
+- `flows/monitor/daily_report.py`(daily-report,18:30):近 7 天 RIIN/SDS 核验(subflow 复用 verify)+商品完整性(`db.get_goods_integrity`)+OCR+失败回顾。
+- `flows/monitor/device_check.py`(device-check,09:00):内存/swap/宿主机磁盘(`/` 挂载在容器 `/host`)/GPU(nvidia-smi)阈值 80% 报警 + Docker 容器异常(docker.sock)+Prefect 失败+OCR/PG 探活。
+- worker 容器已挂载 `/:/host:ro`、`/var/run/docker.sock`、GPU(见 docker-compose.yml),勿移除,否则对应检查降级。
+
 ## 参考文档
 
 - Prefect 入门:https://docs.prefect.org.cn/v3/get-started

flows/monitor/alert_failure.py

@@ -53,12 +53,21 @@ def _flow_run_id(event: dict) -> str | None:
 
 
 def _fetch_run_detail(run_id: str) -> dict:
-    """查 Prefect API 补全 flow run 权威信息,失败返回空 dict。"""
+    """查 Prefect API 补全 flow run 权威信息(含 flow 名),失败返回空 dict。"""
     api_url = os.environ.get("PREFECT_API_URL", "http://localhost:4200/api")
     try:
         resp = httpx.get(f"{api_url}/flow_runs/{run_id}", timeout=10)
         resp.raise_for_status()
-        return resp.json()
+        detail = resp.json()
+        flow_id = detail.get("flow_id")
+        if flow_id:
+            try:
+                fresp = httpx.get(f"{api_url}/flows/{flow_id}", timeout=10)
+                fresp.raise_for_status()
+                detail["_flowName"] = fresp.json().get("name", "")
+            except Exception:  # noqa: BLE001
+                detail["_flowName"] = ""
+        return detail
     except Exception as err:  # noqa: BLE001
         print(f"[wecom-alert] 查询 flow run {run_id} 失败: {err}")
         return {}
@@ -86,7 +95,7 @@ def wecom_alert_failure(event: dict | str | None = None) -> dict:
         flow_name, run_name, state_msg, occurred = "", resource_name, "", ev.get("occurred", "")
         if run_id:
             detail = _fetch_run_detail(run_id)
-            flow_name = detail.get("name", "") or resource_name
+            flow_name = detail.get("_flowName") or resource_name
             run_name = detail.get("name", "")
             state = detail.get("state") or {}
             state_msg = (state.get("message") or "").strip()