docs: record code queue supervision cadence

AGENTS.md

@@ -84,6 +84,7 @@ UniDesk 是一个以主 server 为统一入口的分布式工作平台；本文
 - `docs/reference/user-service-delivery.md`：用户服务默认交付流程、CI 镜像构建与 registry、Baidu Netdisk 主 server 直管微服务样板、dev 自动测试、prod 拉镜像部署和 Decision Center 产品化需求管理规则。
 - `docs/reference/dev-environment.md`：D601 `unidesk-dev` persistent dev 环境、18083 dev frontend proxy、`deploy apply --env dev` 服务范围和 Rust backend-core 只在 D601 编译的边界。
 - `docs/reference/ci.md`：D601 k3s Tekton CI、只读主数据库性能门禁和 CLI 入口规则。
+- `docs/reference/code-queue-supervision.md`：Code Queue 长任务波次的并发窗口、轮询、终态读取、阻塞拆分和 supervisor 边界。
 - `docs/reference/dev-ci-runner.md`：`ci run-dev-e2e` 的 Git 控制 runner、短 launcher、结果目录和 no-CD 边界。
 - `docs/reference/codex-deploy.md`：D601 Code Queue 旧 `codex deploy <commitId>` 入口禁用原因、受控部署边界和后续 CD 目标行为。
 - `reference`：兼容旧路径的符号链接，指向 `docs/reference/`。

docs/reference/code-queue-supervision.md

@@ -52,6 +52,10 @@ Use:
 
 Long-running tasks with fresh trace or heartbeat evidence should normally be left alone. Polling every few minutes is preferred over repeated interrupt/retry cycles.
 
+For broad CI/CD migration waves, use a fixed supervision cadence unless an incident demands faster action. A five-minute poll loop is the default: read `codex queues`, read terminal or suspect task summaries, then either accept, retry, split a blocker, or leave healthy tasks alone. The loop should keep the supervisor doing useful non-overlapping work, such as documentation or issue triage, but that side work must not take over a worker's assigned implementation.
+
+When a task leaves `running` or `judging`, treat the result as unread work until the final response and judge record have been inspected. Only then should the supervisor decide whether to refill the concurrency window.
+
 ## Intervention Rules
 
 Intervene only when there is a clear reason.