@zq-silk/yui 1.0.1 → 1.1.2

package/docs/provider-runtime.zh-CN.md

@@ -117,6 +117,11 @@ Task Session 终止首先取消其确切的未结算输入，并在向 Host 发
 一个活的所有权依赖，而不是虚假静止。这是 OS 进程托管，不是 Task 工作流，也不是关于
 无关外部服务的断言。
 
+macOS 没有子进程 subreaper。owner 会停止仍能通过父子关系或原进程组确认的后代。
+若 Claude 根进程曾 fork，子进程再执行 `setsid` 并被重新托管，就可能避开这两种检查；
+此时 owner 会报告清理未确认，而不会返回成功；普通工具进程曾 fork、但根进程退出后
+无法证明它们都已结束时，也适用这一保守结果。脱离的子进程可能仍需单独检查和终止。
+
 专用进程的 PID/start 身份作为工程控制数据独立于 AgentHost 保留。恢复从不需要旧 Host
 的内存连接。对 Codex，一个单独的元数据/控制客户端可以在不提交另一个 prompt 的情况下
 检查并停止已记录的原生执行；未知接受可以在实际原生静止之后被放弃，而不编造原生 Turn

package/docs/release-workflow.md

@@ -45,10 +45,16 @@ protection and valid exact-identity caches remain normal runtime behavior.
 
 Stable releases use npm `latest`; prereleases use `next`, never the stable
 default. The release workflow builds one runtime archive, verifies it across
-supported Node versions, publishes that exact archive to npm, then creates the
+supported Node versions on Linux x64, macOS x64 and macOS arm64, publishes that exact archive to npm, then creates the
 matching GitHub Release. Every release carries the runtime archive, checksum and
 provenance. A stable tag becomes GitHub latest; a prerelease does not.
 
+Platform jobs compile Yui's native helpers from the same source commit.
+The archive includes every supported helper under `dist/runtime/native/`;
+runtime selection uses the exact OS and architecture, without install scripts
+or downloading executable code on first use. Dependencies such as SQLite
+remain ordinary npm dependencies installed for the consumer's platform.
+
 The GitHub Release job uploads only missing assets to a draft, downloads and
 compares every asset against the tested artifact, and never overwrites a
 published asset. A retry accepts only byte-identical existing evidence. If npm

package/docs/task-delivery.md

@@ -35,9 +35,10 @@ isolation and acceptance checks remain enforced independently.
 
 Stable Project checkouts are read-only references. Task main is a logical
 multi-Project root with independent per-Project Git clones; WorkItem, Review and
-Integration worktrees belong to those Task repositories. For one Project, the Agent's
-normal cwd is its managed Git root; for multiple Projects, the root and native
-additional-directory mechanism expose the explicit Project set.
+Integration worktrees belong to those Task repositories. By default, a Task's native cwd is
+its managed root with zero, one, or multiple Projects. The native
+additional-directory mechanism exposes each bound Project. Binding another
+Project while the Task is active does not move the Task cwd.
 
 An isolated WorkItem has independent worktrees for writable Projects and
 Task-main context for the others. Write scope is explicit and can only be
@@ -367,7 +368,7 @@ This adds neither a retry worker nor another persistent ownership protocol.
 Current Resource records are read strictly: required safety fields, enum values,
 and every active reference must be valid. A malformed record or mismatched
 SQLite/payload identity is reported without supplying defaults, dropping refs,
-or rewriting stored evidence. This enforces the current storage 1.0 record contract.
+or rewriting stored evidence. This enforces the current storage 1.1 record contract.
 
 Failed preparation compensates only its unadopted resources. Standalone Task
 clones use exact clone identity/cleanliness checks before direct deletion;

package/docs/testing/verification-levels.md

@@ -50,13 +50,12 @@ is required.
 ## Permanent core smoke
 
 `npm test` and `npm run test:core` build the checkout and run the maintained suite.
-The stable suite owns only current storage 1.0 and future declared minor
-transitions. It contains no historical conversion implementation, fixture or
-compatibility test.
+The stable suite owns current storage 1.1 and its declared 1.0 minor
+transition. It contains no undeclared historical format compatibility path.
 
 Current coverage includes:
 
-1. Fresh storage 1.0, exact schema/record validation, no initialization over
+1. Fresh storage 1.1, exact schema/record validation, no initialization over
    unknown data, and rejection of old integer formats without mutation.
 2. Exact-version staging, mismatched-target refusal, same-major contiguous
    minor preflight, explicit maintenance-owner identity across handover, and
@@ -99,8 +98,11 @@ do not add prose-matching tests or claim model validation from static checks.
 
 ## CI and release
 
-`ci.yml` builds once and runs core plus one assembled-package normal-path smoke
-on every PR, without another lint or broad regression suite.
+`ci.yml` runs core plus one assembled-package normal-path smoke per supported
+platform (Linux x64, Mac Intel and Apple Silicon) on every PR, without another
+lint or broad regression suite.
+Linux runs on Node 24 and 26; release fresh-install smoke covers Node 20, 22,
+24 and 26 on all three platforms.
 `node scripts/smoke-runtime-package.mjs --assembled .release-stage` exercises
 the actual CLI/Controller/Host/SQLite and isolated tmux, replacing only the
 external Provider with a deterministic fixture. It covers setup, durable input
@@ -111,11 +113,21 @@ and grouped viewers without affecting a similarly named neighboring session.
 The fixture owns a fresh Home and its PATH, installs cleanup before setup,
 and never calls an installed model Agent.
 
-`publish.yml` runs the same smoke against the freshly installed package through
+`publish.yml` builds one archive containing all three native targets and runs
+the same smoke on each platform and supported Node version against that exact
+freshly installed package through
 `YUI_INSTALLED_ROOT`, adding actual npm-bin, dependency, supported Node version,
 artifact and provenance boundaries. This validates runtime integration, not
 real-model behavior. Pure contract and safety tests remain in `test/core`;
 production wiring is exercised here rather than only through mocked ports.
+The native dependency check launches a fixed local program through PTY and
+requires its output and normal exit. Installed-package smoke explicitly checks
+default Doctor's PTY results using dependencies resolved from the consumer.
+On macOS it also tests a disposable copy with a non-executable spawn-helper:
+Doctor must report the selected helper path and permissions without repairing it.
+Doctor's isolated PTY probe waits up to 750 ms for output/exit, with a 1.5-second
+outer native-process limit, and reports elapsed time. Existing Agent/Controller
+diagnostics have their own costs; this is not a one-second whole-Doctor guarantee.
 The package smoke also checks unconditional status identity and update-owned
 resource/identity capture through the assembled package. Real lifecycle children
 stop the exact Controller and restore its captured launch identity while their

package/docs/testing/verification-levels.zh-CN.md

@@ -74,6 +74,8 @@ package-start 检查跟随已安装树中的本地 Skill 引用，包括跨 Role
 
 `ci.yml` 在每个 PR 上构建一次，运行 core 及一个组装包正常链路检查，不重复 lint，
 也不增加宽泛回归套件。
+Linux CI 在 Node 24 和 26 上运行；发布时的全新安装 smoke 覆盖三个平台上的
+Node 20、22、24 和 26。
 `node scripts/smoke-runtime-package.mjs --assembled .release-stage` 经过真实
 CLI/Controller/Host/SQLite 与隔离 tmux，仅用确定性夹具替换外部 Provider。它验证
 setup、输入跨重启持久化及幂等、scratch 激活、原生结果入库、完成后保留会话、
@@ -85,6 +87,12 @@ setup、输入跨重启持久化及幂等、scratch 激活、原生结果入库
 npm bin、依赖、受支持 Node 版本、产物和 provenance 边界。这证明运行时集成，不证明
 真实模型行为。纯契约与安全检查保留在 `test/core`，生产组件组装在这里验证，
 不只依赖模拟端口。
+原生依赖检查必须通过 PTY 启动固定本地程序，确认输出和正常退出。安装包 smoke
+明确断言默认 Doctor 的 PTY 检查成功，并从消费者安装目录解析依赖。macOS 还会
+在可丢弃的依赖副本中移除 spawn-helper 执行权限，验证 Doctor 报出实际 helper
+路径和权限原因且不自动修复。Doctor 的隔离 PTY 探针等待输出／退出最多 750 ms，
+外层原生进程限制为 1.5 秒，并报告耗时；原有 Agent／Controller 检查另有成本，
+这不是整个 Doctor 一秒内返回的保证。
 组装包检查还验证固定身份输出，以及升级侧通过组装包采集资源和精确 Controller 身份，
 并在父进程持有交接锁时，通过真实生命周期子进程停止精确 Controller、恢复其已捕获的
 启动身份。无关调用仍被锁阻止，锁保持由父进程持有，持久输入不变；

package/i18n/README.zh-CN.md

@@ -5,7 +5,7 @@
 [![Core CI](https://github.com/zhangqian-silk/yui/actions/workflows/ci.yml/badge.svg?branch=master)](https://github.com/zhangqian-silk/yui/actions/workflows/ci.yml)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](../LICENSE)
 ![Node](https://img.shields.io/badge/node-20%20%7C%2022%20%7C%2024-brightgreen.svg)
-![Platform](https://img.shields.io/badge/platform-Linux%20x64%20%28glibc%29-blue.svg)
+![Platform](https://img.shields.io/badge/platform-Linux%20x64%20%28glibc%29%20%7C%20macOS%20x64%2Farm64-blue.svg)
 [![PRs welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](#参与开发)
 
 让 Agent 持续推进你的任务，而不只是回答一轮对话。
@@ -30,16 +30,21 @@ Yui 是面向编程 Agent 的本地控制面。你只需用自然语言把目标
   Web 仅本地回环，包含只读视图和经认证的用户控制。
 - **默认隔离** —— 仓库改动发生在受管 Git worktree 中，稳定 checkout 保持只读。
 
-> **状态：** 1.0.0，采用纯净的存储 1.0 基线。默认更新只执行同一存储
+> **状态：** 1.1.0，当前存储版本为 1.1，已声明从 1.0 基线升级。默认更新只执行同一存储
 > 主版本内已声明的小版本升级；其他存储身份会在不修改 Home 的前提下被拒绝。
 
 [快速开始](#快速开始) · [通过对话管理工作](#通过对话管理工作) · [架构](#架构) · [设计原则](#设计原则)
 
 ## 快速开始
 
-需要 Linux x64 / glibc、Git、tmux，以及 Node.js `^20.17.0`、`^22.9.0` 或
-`^24.0.0`。最简单的方式是先安装 Codex CLI 或 Claude Code CLI，并确保它
-已经可以使用你自己的账号正常工作。Yui 负责协调 Agent，不提供模型访问额度。
+npm 包包含 Linux x64、Mac Intel 和 Apple Silicon 的 Yui 预编译程序。
+同一份包在三个平台验证后发布，运行时只选择对应平台的程序。
+
+需要 Linux x64 / glibc 或 macOS（x64 或 Apple Silicon）、Git、tmux，以及
+Node.js `^20.17.0`、`^22.9.0`、`^24.0.0` 或 `^26.0.0`。macOS 上可用
+`brew install tmux` 安装 tmux。最简单的方式是先安装 Codex CLI 或 Claude
+Code CLI，并确保它已经可以使用你自己的账号正常工作。Yui 负责协调 Agent，
+不提供模型访问额度。
 
 Yui 传递 Claude 的认证环境并保留原生配置目录，由 Claude 自身按本地配置
 选择 API key 或登录方式。更换 Session 不会重置登录或初始化记录。首次运行时，
@@ -64,10 +69,18 @@ yui setup
 > 我已经安装了 Yui。请在交互式终端中帮我执行 `yui setup`，选择可用的
 > Agent，并用 `yui doctor` 检查结果。遇到账号或需要我决定的配置时问我。
 
-Setup 会配置与你对话的 Operator 和默认任务 Leader，并启动本地 Controller。
+Setup 会配置与你对话的 Operator 和默认任务 Leader，并启动本地 Controller，
+以及这些活跃 Role 需要的 Codex 共享 daemon。
 先用这两个角色即可开始，Worker、Reviewer 可以之后再配置。如果 Agent
 没有操作交互式终端的能力，就自己运行 setup；它只负责初始配置。
 
+之后启动时运行 `yui start`。它会按需启动 Controller；只有活跃 Role 使用
+Codex 时才会启动 Codex 共享 daemon。在 macOS 上，如果安装了 Codex App，
+它还会配置 App 下次启动时连接该 daemon。先运行 `yui start` 再打开 App；
+如果 App 已经打开，按 Yui 的提示退出并重新打开。`yui doctor` 可以检查
+实时连接。`yui controller start` 只启动 Controller。Linux 不需要 App 配置；
+没有活跃 Codex Role 时不会启动 Codex daemon。
+
 ### 3. 开始对话
 
 ```sh

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@zq-silk/yui",
-  "version": "1.0.1",
+  "version": "1.1.2",
   "description": "Local control plane for long-running native agent CLI sessions backed by tmux.",
   "license": "MIT",
   "private": false,
@@ -19,16 +19,15 @@
     "LICENSE"
   ],
   "engines": {
-    "node": "^20.17.0 || ^22.9.0 || ^24.0.0"
+    "node": "^20.17.0 || ^22.9.0 || ^24.0.0 || ^26.0.0"
   },
   "os": [
-    "linux"
+    "linux",
+    "darwin"
   ],
   "cpu": [
-    "x64"
-  ],
-  "libc": [
-    "glibc"
+    "x64",
+    "arm64"
   ],
   "keywords": [
     "yui",
@@ -50,7 +49,7 @@
     "@xterm/addon-fit": "^0.11.0",
     "@xterm/xterm": "^6.0.0",
     "better-sqlite3": "^12.11.1",
-    "node-pty": "^1.1.0",
+    "node-pty": "1.2.0-beta.15",
     "smol-toml": "1.8.0",
     "ws": "^8.21.1"
   }

package/skills/yui-leader/SKILL.md

@@ -69,6 +69,14 @@ The DB Brief is the current summary, not a full plan or a history of messages.
 After meaningful progress, use `task brief update` with only intended fields;
 read `task event list` before deliberately restoring an older value. Keep
 references to substantial plans, prototypes and reports in that summary.
+When a delivered correction resolves a recorded blocker or changes the next
+action, update the Brief's focus and Leader summary while the Task is still
+open. Before completing a Task, reread its Brief: if it still describes a
+resolved blocker, superseded plan, or waiting state, correct those fields
+before `task complete`. The completion summary is the terminal conclusion;
+do not copy it into Brief merely to create a second completion record. Brief
+maintenance is not a new completion gate, and routine unchanged steps do not
+need an update.
 Decisions contain the actual decision, reason and necessary boundaries, not
 the entire proposal. Label recommendations as recommendations, not user
 decisions.

package/skills/yui-leader/references/execution.md

@@ -409,7 +409,10 @@ Before ending authorized execution:
    with `yui task run show`, read each original result in full, and make the
    next decision.
 2. Persist actual WorkItem lifecycle and material Brief, Decision, Milestone,
-   Message, or Knowledge changes.
+   Message, or Knowledge changes. If an accepted result or authorized
+   correction has cleared a blocker still stated as current in Brief, update
+   that working summary before completing or yielding. Keep the original
+   failure and correction in their durable records.
 3. Choose one truthful outcome: continue through an owned native child, complete
    the Task, create a justified InputRequest, or leave the active Task waiting
    for a real durable event.