Release beta 1.0.4 update settlement

AGENTS.md

@@ -24,6 +24,22 @@ These rules are mandatory for every automated change in this repository.
   or adaptive restart-stability gates.
 - If a deployment gate fails, verify rollback state before retrying.
 
+## Mandatory Release Settlement
+
+- Every successful production update must run `jwebot_release_update.py`
+  after cutover gates pass and before the release is declared complete.
+- Each release grants exactly 200 Mora to every account that exists when the
+  settlement starts. The grant and account movement are idempotent by version.
+- The first bullet in each release's `CHANGELOG.md` section must be one concise,
+  user-facing sentence suitable for the group update announcement.
+- After crediting accounts, send the version, first changelog bullet, and
+  `Mora x200` compensation notice through the guarded group sender.
+- Never blindly retry an uncertain announcement. Verify the group visually,
+  then use `--retry-uncertain-announcement` only when the prior message did not
+  land.
+- A production release is incomplete until compensation, announcement,
+  SQLite integrity, empty queues, Edge health, and service stability pass.
+
 ## Configuration And Access
 
 - Real profiles live under `~/.config/jwebot/`, never in the repository.

CHANGELOG.md

@@ -1,5 +1,38 @@
 # Changelog
 
+## beta 1.0.4.1016 - 2026-10-16
+
+- 新增版本更新补偿流程：每次生产更新后向所有既有账户发放 200 摩拉，并在目标群发布更新公告。
+- Add a version-idempotent post-release settlement tool that creates an online
+  SQLite backup, grants Mora through the production account schema, records a
+  version-labelled movement, and tracks announcement delivery independently.
+- Run release settlement only after coordinated cutover and rollback gates are
+  disarmed, keeping a healthy runtime active when GUI delivery is uncertain.
+- Require the first changelog bullet to be a concise user-facing announcement
+  summary and require visual confirmation before explicitly retrying an
+  uncertain announcement.
+
+## beta 1.0.3.1016 - 2026-10-16
+
+- Rename the virtual currency user-facing copy to Mora (`摩拉`) while keeping
+  the existing `/金币`, `/金币排行`, and `/Taffy金币` commands as aliases.
+- Add recent account movement details to `/摩拉` and `/金币`, covering the
+  latest transfer, loan, repayment, lottery, draw, and blackjack net changes.
+- Suppress short-window semantic duplicate OCR events so one visible command
+  bubble cannot trigger the same local reply multiple times under new event IDs.
+
+## beta 1.0.2.1016 - 2026-10-16
+
+- Cache OCR-negative visible bubbles and reuse stable visual signatures as the
+  OCR cache key so unreadable or lightly re-rendered bubbles do not trigger
+  repeated 10+ second RapidOCR passes during active chat windows.
+
+## beta 1.0.1.1016 - 2026-10-16
+
+- Limit RapidOCR ONNXRuntime intra/inter operation threads by default so the
+  adaptive OCR recorder no longer saturates CPU after restart and slow recovery
+  windows do not stretch local command detection into 10+ second scans.
+
 ## beta 1.0.0.1016 - 2026-10-16
 
 - Promote JWeBot from alpha to beta after the local, staging, privacy, and

README.md

@@ -8,12 +8,12 @@ JWeBot 是运行于 Linux 的单群微信自动回复系统。它通过本地 OC
 观察选定的微信窗口，将消息写入持久化 Edge 账本，调用经过审核的命令插件，
 并通过带视觉确认的 GUI 发送链路回复。
 
-当前版本：`beta 1.0.0.1016`，更新日期：2026-10-16。
+当前版本：`beta 1.0.4.1016`，更新日期：2026-10-16。
 
 1.0 Beta 收口了当前稳定功能：本地命令插件、人物与上下文索引、
-Steam 状态与时长查询、虚拟金币游戏、多人彩票、21 点、CS Major
+Steam 状态与时长查询、虚拟摩拉游戏、多人彩票、21 点、CS Major
 自走棋、生产部署与回滚门禁。CS Major 自走棋的选手、教练、叙事和规则
-已经外置为包内 JSON，虚拟金币支持转账、贷款、还款和群内排行榜。
+已经外置为包内 JSON，虚拟摩拉支持转账、贷款、还款和群内排行榜。
 
 完整命令见[命令目录](docs/command-catalog.md)。历史版本记录见
 [CHANGELOG](CHANGELOG.md)。
@@ -82,12 +82,13 @@ PATH="$PWD/.venv/bin:$PATH" ./tools/check-local.sh
 /功能列表
 /上下文 [100|200|300]（/context）
 /人物列表 [页码]（/people）；/人物 [名称或别名]（/person）
-/金币（/wallet）；/抽奖 [金额]（/draw）
+/摩拉（兼容 /金币、/wallet）；/抽奖 [金额]（/draw）
-/大富翁榜（/金币排行）
+/大富翁榜（/摩拉排行，兼容 /金币排行）
 /转账 对方 金额（/transfer name amount）
 /贷款 [金额]（/loan amount）；/还款 [金额|全部]（/repay）
 /彩票 开始|报名|当前状态|开奖|取消（/lottery）
 /21点 开始 [1～500]；开局后 /21点 要牌|停牌|状态|结束
+/21点 多人 启动；/21点 加入；/21点 多人 开始；/21点 放200
 /游戏记录 [页码]（/gamehistory）
 /游戏帮助（/games）
 /自走棋开始|报名|开赛|状态|档案
@@ -188,7 +189,9 @@ sudo ./deploy/scripts/deploy-jwebot-production.sh /path/to/JWeBot jwebot
 ```
 
 部署必须通过微信窗口就绪、队列清空、Edge 健康、SQLite 完整性和自适应
-录制服务重启稳定性检查。旧 OCR 服务保留安装，但保持禁用。
+录制服务重启稳定性检查。切换成功后，部署器按版本幂等地向每个既有账户
+发放 200 摩拉、写入账户流水，并使用当前 changelog 的第一条摘要发布群公告。
+旧 OCR 服务保留安装，但保持禁用。
 
 ### 仓库结构
 
@@ -262,7 +265,7 @@ visually confirmed GUI bridge.
 Current version:
 
 ```text
-beta 1.0.0.1016
+beta 1.0.4.1016
 ```
 
 Updated: 2026-10-16.
@@ -451,7 +454,9 @@ sudo ./deploy/scripts/deploy-jwebot-production.sh /path/to/JWeBot jwebot
 
 The deployer refuses cutover unless the ready-window, empty-queue, Edge health,
 SQLite integrity, and adaptive restart-stability gates pass. Legacy OCR units
-remain installed but disabled.
+remain installed but disabled. After cutover, version-idempotent release
+settlement grants 200 Mora to every existing account, records the movement,
+and sends the first changelog bullet as the guarded group announcement.
 
 ## Repository Map
 

VERSION

@@ -1 +1 @@
-beta 1.0.0.1016
+beta 1.0.4.1016

deploy/scripts/deploy-jwebot-production.sh

@@ -421,6 +421,7 @@ test -x "${runtime_python}"
 for path in \
   "${source_dir}/wechat_local_ocr.py" \
   "${source_dir}/wechat-gui-bridge" \
+  "${source_dir}/jwebot_release_update.py" \
   "${source_dir}/deploy/scripts/install-wechat-health-watch.sh" \
   "${source_dir}/deploy/systemd/system/wechat-local-ocr-recorder.service" \
   "${source_dir}/deploy/systemd/system/wechat-local-ocr-responder.service" \
@@ -430,7 +431,9 @@ do
 done
 bash -n "${source_dir}/wechat-gui-bridge"
 sh -n "${source_dir}/deploy/scripts/install-wechat-health-watch.sh"
-"${runtime_python}" -m py_compile "${source_dir}/wechat_local_ocr.py"
+"${runtime_python}" -m py_compile \
+  "${source_dir}/wechat_local_ocr.py" \
+  "${source_dir}/jwebot_release_update.py"
 preflight_rendered_dir=$(mktemp -d)
 render_root_configuration "${preflight_rendered_dir}"
 systemd-analyze verify \
@@ -619,6 +622,17 @@ success=1
 armed=0
 trap - EXIT HUP INT TERM
 
+jwebot_log "release settlement starting"
+release_update_output=$(
+  /usr/sbin/runuser -u "${target_user}" -- /usr/bin/env \
+    HOME="${target_home}" \
+    XDG_RUNTIME_DIR="${user_runtime}" \
+    DBUS_SESSION_BUS_ADDRESS="unix:path=${user_runtime}/bus" \
+    "${runtime_python}" "${target_runtime}/jwebot_release_update.py"
+)
+printf '%s\n' "${release_update_output}"
+jwebot_log "release settlement ready"
+
 jwebot_log "deployment ready version=$(cat "${source_dir}/VERSION")"
 jwebot_log "backup active path=${backup_dir}"
 jwebot_log "legacy OCR inactive"

deploy/scripts/install-wechat-health-watch.sh

@@ -27,6 +27,7 @@ adaptive_script_name=wechat_adaptive_recorder.py
 adaptive_bridge_name=wechat_adaptive_gui_bridge.py
 edge_script_name=jwebot_edge.py
 edge_delivery_name=jwebot_edge_delivery.py
+release_update_name=jwebot_release_update.py
 service_name=wechat-health-watch.service
 timer_name=wechat-health-watch.timer
 adaptive_service_name=wechat-ocr-adaptive-recorder.service
@@ -209,6 +210,7 @@ rollback() {
         restore_file "$adaptive_bridge_name" "$target_dir/$adaptive_bridge_name" 0700
         restore_file "$edge_script_name" "$target_dir/$edge_script_name" 0700
         restore_file "$edge_delivery_name" "$target_dir/$edge_delivery_name" 0700
+        restore_file "$release_update_name" "$target_dir/$release_update_name" 0700
         for package_name in \
             jwebot_config \
             jwebot_sdk \
@@ -246,6 +248,7 @@ for path in \
     "$runtime_source_dir/$adaptive_bridge_name" \
     "$runtime_source_dir/$edge_script_name" \
     "$runtime_source_dir/$edge_delivery_name" \
+    "$runtime_source_dir/$release_update_name" \
     "$user_unit_source_dir/$service_name" \
     "$user_unit_source_dir/$timer_name" \
     "$user_unit_source_dir/$adaptive_service_name" \
@@ -289,7 +292,8 @@ for name in \
     "$adaptive_script_name" \
     "$adaptive_bridge_name" \
     "$edge_script_name" \
-    "$edge_delivery_name"
+    "$edge_delivery_name" \
+    "$release_update_name"
 do
     if [ -f "$target_dir/$name" ]; then
         cp -a "$target_dir/$name" "$backup_dir/$name"
@@ -342,6 +346,9 @@ install -m 0700 "$runtime_source_dir/$adaptive_script_name" "$target_dir/$adapti
 install -m 0700 "$runtime_source_dir/$adaptive_bridge_name" "$target_dir/$adaptive_bridge_name"
 install -m 0700 "$runtime_source_dir/$edge_script_name" "$target_dir/$edge_script_name"
 install -m 0700 "$runtime_source_dir/$edge_delivery_name" "$target_dir/$edge_delivery_name"
+install -m 0700 \
+    "$runtime_source_dir/$release_update_name" \
+    "$target_dir/$release_update_name"
 rm -rf \
     "$target_dir/jwebot_config" \
     "$target_dir/jwebot_sdk" \
@@ -395,7 +402,8 @@ for name in \
     "$adaptive_script_name" \
     "$adaptive_bridge_name" \
     "$edge_script_name" \
-    "$edge_delivery_name"
+    "$edge_delivery_name" \
+    "$release_update_name"
 do
     "$runtime_python" -m py_compile "$target_dir/$name"
 done

docs/command-catalog.md

@@ -1,6 +1,6 @@
 # JWeBot Command Catalog
 
-Current baseline: `beta 1.0.0.1016`.
+Current baseline: `beta 1.0.1.1016`.
 
 This catalog records the reviewed commands shipped with the built-in local
 extensions. Runtime ownership is frozen at startup from the generated
@@ -41,8 +41,8 @@ descriptor inventory; plugin code delegates to reviewed service profiles.
 
 | Command | Alias | Purpose |
 |---|---|---|
-| `/金币` | `/wallet` | Initialize or show the current person's virtual coin account. |
+| `/摩拉` | `/金币`, `/wallet` | Initialize or show the current person's Mora account. |
-| `/大富翁榜 [人数]` | `/金币排行`, `/金币排行榜`, `/richlist` | Show the group-scoped coin leaderboard. |
+| `/大富翁榜 [人数]` | `/摩拉排行`, `/摩拉排行榜`, `/金币排行`, `/金币排行榜`, `/richlist` | Show the group-scoped Mora leaderboard. |
 | `/转账 对方 金额` | `/transfer name amount` | Transfer virtual coins to an initialized group account. |
 | `/贷款 [金额]` | `/loan amount`, `/借款` | Borrow up to 500 virtual coins. |
 | `/还款 [金额|全部]` | `/repay [amount|all]` | Repay an active virtual coin loan. |
@@ -57,6 +57,10 @@ descriptor inventory; plugin code delegates to reviewed service profiles.
 | `/21点 停牌` | `/blackjack stand` | Stand and settle against the dealer. |
 | `/21点 状态` | `/blackjack status` | Show the current blackjack round. |
 | `/21点 结束` | `/blackjack end` | End the current blackjack round. |
+| `/21点 多人 启动` | - | Open a Taffy-hosted blackjack table and auto-seat the host. |
+| `/21点 加入` | - | Join the current blackjack table before betting starts. |
+| `/21点 多人 开始` | - | Host-only close joining and start ordered betting prompts. |
+| `/21点 放金额` | - | Place the prompted table stake, 1-10000 coins. |
 | `/游戏记录 [页码]` | `/gamehistory` | Show indexed personal game history. |
 | `/游戏帮助` | `/games` | Show local game rules. |
 

docs/configuration-guide.md

@@ -90,6 +90,15 @@ Steam uses `JWEBOT_EDGE_STEAM_ID64` and
 - Host timeout and fallback;
 - GUI sender timeout.
 
+The OCR runtime also accepts optional environment overrides for constrained
+hosts:
+
+- `WECHAT_OCR_ONNX_INTRA_OP_THREADS` defaults to `1`;
+- `WECHAT_OCR_ONNX_INTER_OP_THREADS` defaults to `1`.
+
+These values limit RapidOCR's ONNXRuntime CPU thread pools so service restarts
+do not temporarily saturate the host while the OCR cache is cold.
+
 ### Debug
 
 `components/debug.toml` generates:

docs/cs-major-autochess-design.md

@@ -48,7 +48,7 @@
 每位玩家开局拥有：
 
 - 生命值：`100`
-- 金币：`10`
+- 摩拉：`10`
 - 教练：赛前第一轮选秀决定
 - 初始选手：赛前第二、三轮选秀决定，系统随后补齐阵容
 - 阵容位：固定 `5 个选手位 + 1 个教练位`，比赛时必须全部上场
@@ -206,7 +206,7 @@
 
 ### 7.2 教练获取
 
-教练由第一轮公共选秀决定，本届赛事内不能用金币升级教练。不同教练
+教练由第一轮公共选秀决定，本届赛事内不能用摩拉升级教练。不同教练
 自带等级、属性倾向和全队加成。
 高费选手的商店概率随赛事进程推进，不再绑定教练等级。
 
@@ -270,7 +270,7 @@
 
 - 第一轮教练选秀展示固定 10 名教练；第二、三轮选手选秀展示 6 名选手。
 - 训练赛和正赛运营阶段默认展示一组 6 人商店。
-- 玩家可支付金币执行刷新。
+- 玩家可支付摩拉执行刷新。
 - 玩家可锁定本轮 6 人，使下一轮仍保留同一组候选。
 
 ### 8.4 顺位
@@ -294,7 +294,7 @@
 ## 10. 训练赛与赛事挑战
 
 训练赛不再是阈值型 PvE。它是正式小组赛前的多轮真实对阵，用来给玩家
-检验阵容、获取金币并自然插入 D 牌窗口。媒体日和赞助挑战仍作为赛事内
+检验阵容、获取摩拉并自然插入 D 牌窗口。媒体日和赞助挑战仍作为赛事内
 插曲出现，但只提供资源变动，不淘汰玩家。
 
 ### 10.1 活动类型
@@ -308,11 +308,11 @@
 
 训练赛按参赛队伍生成 round-robin 轮次。四队测试局会有三轮训练赛，
 每队在训练阶段遇到另外三支队伍；奇数队伍时轮空队会与本地陪练队热身。
-训练赛只发放金币，不记录小组胜负，也不扣生命。
+训练赛只发放摩拉，不记录小组胜负，也不扣生命。
 
 每轮训练赛结束后进入运营窗口：
 
-- `/刷新` 即 D 牌，消耗金币换一组 6 人商店；
+- `/刷新` 即 D 牌，消耗摩拉换一组 6 人商店；
 - `/选1～/选6` 买入商店选手，重复卡自动累积星级；
 - `/锁定` 保留当前商店；
 - `/上阵` 调整 5 人首发；
@@ -321,10 +321,10 @@
 ### 10.3 训练赛结算逻辑
 
 训练赛使用与正式比赛相同的单图 MR15 逐回合模拟，只是赛制为 BO1。
-胜方获得更多金币，负方也有保底金币，保证测试局能继续运营。
+胜方获得更多摩拉，负方也有保底摩拉，保证测试局能继续运营。
 
 媒体日和赞助挑战根据阵容完整度、教练、战术和本地确定性随机因子发放
-金币，但播报不展示“目标/发挥”这类计算日志，只写剧情和奖励。
+摩拉，但播报不展示“目标/发挥”这类计算日志，只写剧情和奖励。
 
 ## 11. PvP 赛制
 
@@ -618,7 +618,7 @@ Jacky 选择了 zonic！
 - AI 测试战队加入
 - 赛制与种子公布
 - 三轮选秀候选池、选择结果和选秀完成
-- 每轮训练赛对阵、比分、金币和运营提示
+- 每轮训练赛对阵、比分、摩拉和运营提示
 - 媒体日或赞助挑战
 - 每场 BO3/BO5 对阵与结果
 - 每张加时地图

docs/group-lottery-0.6.3.md

@@ -4,7 +4,7 @@
 
 1. 任意群友发送 `/彩票 开始`。发起人是本轮主持人。
 2. 每个人发送 `/彩票 报名 3 4 7 100`，表示选择 3、4、7，下注 100
-   金币。也可简写为 `/彩票347100`；省略金额时默认下注 10。
+   摩拉。也可简写为 `/彩票347100`；省略金额时默认下注 10。
 3. 发送 `/彩票 当前状态` 查看本轮主持人、报名人数、每个人的号码和下注。
 4. 主持人发送 `/彩票 开奖`。机器人从 0～9 中抽取三个互不重复的数字，
    统一排序后结算。
@@ -45,7 +45,7 @@
 | 2 个 | 120% |
 | 3 个 | 300% |
 
-金币只使用整数；百分比计算的小数部分向下取整。例如下注 100，命中两个
+摩拉只使用整数；百分比计算的小数部分向下取整。例如下注 100，命中两个
 返还 120，净变化为 `+20`。
 
 ## 个人局号与记录

docs/local-plugins-0.6.1.md

@@ -7,8 +7,8 @@ OpenClaw；中文、英文命令对应同一实现，英文不区分大小写，
 | 插件 | 中文指令 | 英文指令 |
 |---|---|---|
 | context-summary | `/上下文 [100|200|300]` | `/context [100|200|300]` |
-| single-player-games | `/金币` | `/wallet` |
+| single-player-games | `/摩拉`（兼容 `/金币`） | `/wallet` |
-| coin-ranking | `/大富翁榜`、`/金币排行` | `/richlist` |
+| coin-ranking | `/大富翁榜`、`/摩拉排行`（兼容 `/金币排行`） | `/richlist` |
 | coin-transfer | `/转账 对方 金额` | `/transfer name amount` |
 | coin-loan | `/贷款 [金额]`、`/还款 [金额\|全部]` | `/loan amount`、`/repay [amount\|all]` |
 | single-player-games | `/抽奖 [金额]` | `/draw [金额]` |
@@ -23,7 +23,7 @@ OpenClaw；中文、英文命令对应同一实现，英文不区分大小写，
 命令所有权来自 `descriptors.py`，由同步工具生成三套受审核 SDK adapter。
 每个 adapter 只提交命名 service-profile 请求，由 Kernel 执行数据库操作。
 注册在启动期完成，运行期冻结。`/余额` 保持查询提供商余额；
-游戏虚拟余额使用 `/金币`。
+游戏虚拟余额使用 `/摩拉`，`/金币` 保留为兼容别名。
 
 ## 上下文总结
 
@@ -43,7 +43,7 @@ OpenClaw；中文、英文命令对应同一实现，英文不区分大小写，
 
 当前 OCR 输入只有显示名，没有微信原生账号 ID。Kernel 将昵称或别名
 解析到私有配置中的 canonical_name，再以“当前群＋规范人物名”计算
-person_key。相同人物的已配置别名共用金币和牌局，其他群使用独立账户。
+person_key。相同人物的已配置别名共用摩拉和牌局，其他群使用独立账户。
 没有配置的显示名使用独立临时身份；未识别到发送人的消息不能操作账户。
 
 `/人物` 默认查看本人；指定参数时只查询已配置人物。输出包括展示名、
@@ -53,27 +53,27 @@ person_key。相同人物的已配置别名共用金币和牌局，其他群使
 
 昵称识别不是账号身份认证。修改 canonical_name 会改变当前人物作用域；
 应优先追加别名而不是重命名。除 `/转账` 外，用户不能通过游戏命令
-指定他人的账户；转账只支持同群虚拟金币账户，收款人需先发送 `/金币`
+指定他人的账户；转账只支持同群虚拟摩拉账户，收款人需先发送 `/摩拉`
-初始化。`/贷款` 和 `/还款` 只作用于本人账户。虚拟金币不能充值、
+或 `/金币` 初始化。`/贷款` 和 `/还款` 只作用于本人账户。虚拟摩拉不能充值、
 赠币或兑换真实资产。
 
-## 虚拟金币与游戏管理
+## 虚拟摩拉与游戏管理
 
-每个群内人物首次使用账户时一次性获得 1000 虚拟金币。再次查询、
+每个群内人物首次使用账户时一次性获得 1000 虚拟摩拉。再次查询、
 重启服务或切换已配置别名都不会重复领取。抽奖下注为 2～100，彩票和
 21 点下注为 1～500；余额不足或参数非法不扣款。
 
-`/转账 对方 金额` 会把本人的虚拟金币转给同群已初始化账户，英文为
+`/转账 对方 金额` 会把本人的虚拟摩拉转给同群已初始化账户，英文为
 `/transfer name amount`。金额范围 1～100000；不允许给自己转账。
 转账、响应记录和重试去重写在同一个 ledger 事务内，同一 event_id
 不会重复扣款。
 
-`/贷款 [金额]` 会给本人账户借入 1～500 金币；同一账户同一时间最多
+`/贷款 [金额]` 会给本人账户借入 1～500 摩拉；同一账户同一时间最多
-一笔未还贷款。`/还款 [金额|全部]` 从当前金币余额中扣除并减少未还
+一笔未还贷款。`/还款 [金额|全部]` 从当前摩拉余额中扣除并减少未还
 贷款，借款、还款、响应记录和重试去重同样写在一个 ledger 事务内。
 
-`/大富翁榜` 或 `/金币排行 [人数]` 只读取当前群已有金币账户，不会创建
+`/大富翁榜` 或 `/摩拉排行 [人数]`（兼容 `/金币排行`）只读取当前群已有摩拉账户，不会创建
-调用者账户。榜单按当前金币余额降序排列，人数范围 1～20；账户存在
+调用者账户。榜单按当前摩拉余额降序排列，人数范围 1～20；账户存在
 未还贷款时会在名字后标注贷款本金。
 
 抽奖默认下注 10：5% 概率返还五倍下注，20% 返还两倍，其余无返还。
@@ -81,7 +81,7 @@ alpha 0.6.3 将旧的个人即时彩票替换为群内多人轮次。三位数
 下注范围 1～500，命中 1/2/3 个分别返还 50%/120%/300%。完整流程见
 [多人彩票与游戏记录](group-lottery-0.6.3.md)。
 
-21 点默认下注 50、范围 1～500。`/21点 开始 100` 表示拿 100 金币作为本局下注进场，
+21 点默认下注 50、范围 1～500。`/21点 开始 100` 表示拿 100 摩拉作为本局下注进场，
 等价的 OCR 紧凑形式是 `/21点开始100`。开局会扣除下注并发两张牌，
 然后根据牌面单独发送 `/21点 要牌` 或 `/21点 停牌`；开局与操作写在
 同一条消息时不会扣款，而会返回分步提示。牌局使用洗牌后的单副 52 张牌，
@@ -90,6 +90,12 @@ alpha 0.6.3 将旧的个人即时彩票替换为群内多人轮次。三位数
 仅针对本人牌局。发牌、状态和结算回复以当前群内人物的配置展示名开头，
 例如 `jacky #1：你的牌 ...`；别名解析到同一人物时沿用相同展示名、
 账户、局号和游戏记录。
+
+多人 21 点使用 `/21点 多人 启动` 创建 Taffy 牌桌，开桌者自动入座。
+其他玩家在下注前用 `/21点 加入` 进入当前牌桌；主持人发送 `/21点 多人 开始`
+后，Taffy 会按座位顺序提示每个人 `/21点 放200` 放筹码，范围 1～10000。
+所有玩家下注后统一发牌，轮到本人时用 `/21点 要牌` 或 `/21点 停牌` 操作；
+结算按座位顺序列出每个人的下注、返还、净变化和当前摩拉。
 自然 21 点立即结算，返还 2.5 倍下注向下取整；普通胜利返还两倍，
 平局返还下注，失败不返还。每人最多一局；牌局进行中不接受另一笔
 抽奖或 21 点开局，多人彩票报名仍可独立进行。
@@ -100,7 +106,7 @@ alpha 0.6.3 将旧的个人即时彩票替换为群内多人轮次。三位数
 
 Kernel 用一个 `BEGIN IMMEDIATE` 事务同时保存：
 
-1. `player_accounts` 中的金币、牌局和胜负统计；
+1. `player_accounts` 中的摩拉、牌局和胜负统计；
 2. `player_actions` 中的事件、命令及响应。
 
 同一群中的同一 event_id 只结算一次，重试返回原响应。相同事件若变更

docs/project-rules.md

@@ -18,6 +18,9 @@ exception.
   - every package `__version__`;
   - `README.md`;
   - `CHANGELOG.md`.
+- The first bullet under each release heading in `CHANGELOG.md` must be one
+  concise, user-facing sentence. It is the canonical group announcement
+  summary.
 - Do not mark production complete merely because code was committed. Production
   is complete only after staging, cutover, acceptance, observation, and backup
   restore checks pass.
@@ -63,6 +66,26 @@ Before touching production runtime files on production host:
   SQLite gates.
 - If any gate fails, confirm rollback state before retrying.
 
+## Mandatory Post-Release Settlement
+
+- After cutover and rollback gates pass, run the installed
+  `jwebot_release_update.py`.
+- The settlement grants exactly 200 Mora to every account that exists at the
+  start of that release settlement and appends a version-labelled account
+  movement.
+- Compensation uses the full release string as its idempotency key. Rerunning
+  the same release must not grant Mora twice.
+- The tool sends one guarded announcement containing the version, the first
+  changelog bullet, and the `Mora x200` compensation notice.
+- Announcement state is tracked separately from compensation. A missing or
+  uncertain GUI receipt must never roll back a healthy runtime or trigger
+  another grant.
+- If delivery is uncertain, inspect the configured group before using
+  `--retry-uncertain-announcement`; never retry blindly.
+- Production is not complete until the settlement reports success, SQLite
+  `quick_check` is `ok`, queues are empty, and service restart counts remain
+  stable.
+
 ## Backup And Restore
 
 - Pre-deploy and post-deploy baselines must include runtime files, systemd

docs/runbooks/production-deployment.md

@@ -73,7 +73,12 @@ The deployer prints an eight-stage progress bar and:
 - leaves legacy OCR disabled;
 - verifies Edge health and adaptive restart stability;
 - writes and verifies backup SHA-256 checksums;
-- rolls back files, state, and unit topology on failure.
+- rolls back files, state, and unit topology on cutover failure;
+- after rollback is disarmed, grants each existing account 200 Mora, records a
+  version-labelled movement, and sends the guarded group update announcement.
+
+The first bullet in the current `CHANGELOG.md` release section is the canonical
+one-sentence announcement summary. Keep it concise and user-facing.
 
 ## Post-Cutover
 
@@ -104,6 +109,28 @@ PY
 Player-visible acceptance must use deliberate synthetic messages in the
 configured group and confirm exactly one ACK and one final response.
 
+The deployer also runs the installed release settlement tool. Its successful
+JSON result must report the expected account count and announcement delivery.
+The operation is idempotent by the full release string, so rerunning the same
+command never grants Mora twice:
+
+```sh
+/opt/jwebot/venv/bin/python \
+  ~/.local/lib/wechat-health-watch/jwebot_release_update.py
+```
+
+If it reports `AnnouncementUncertainError`, inspect the configured group first.
+Only when the announcement is confirmed absent may the operator retry:
+
+```sh
+/opt/jwebot/venv/bin/python \
+  ~/.local/lib/wechat-health-watch/jwebot_release_update.py \
+  --retry-uncertain-announcement
+```
+
+Do not repeat the full deployment to repair an uncertain announcement. The
+runtime cutover and version-keyed compensation may already be complete.
+
 ## Rollback
 
 If any gate fails:

docs/versioning.md

@@ -14,9 +14,9 @@ Rules:
 - `X.Y.Z` is the project version.
 - `MMDD` is the delivery date.
 - Python package versions use PEP 440 pre-release form. For beta builds,
-  `beta 1.0.0.1016` maps to `1.0.0b1+1016`.
+  `beta 1.0.1.1016` maps to `1.0.1b1+1016`.
 - Normal small updates increment only the patch version, for example
-  `beta 1.0.0.1016` to `beta 1.0.1.1016`.
+  `beta 1.0.1.1016` to `beta 1.0.2.1016`.
 - Minor, major, or stable-release changes require explicit user review and
   milestone approval.
 - Local repository migration does not imply a production deployment.

extensions/jwebot/coin-ranking/README.md

@@ -10,6 +10,8 @@ Help:
 
 Registered commands:
 
+- `/摩拉排行`
+- `/摩拉排行榜`
 - `/金币排行`
 - `/金币排行榜`
 - `/大富翁榜`

extensions/jwebot/coin-ranking/plugin.toml

@@ -8,7 +8,7 @@ entrypoint = "coin_ranking:plugin"
 type = "command"
 
 [capabilities]
-commands = ["/金币排行", "/金币排行榜", "/大富翁榜", "/richlist"]
+commands = ["/摩拉排行", "/摩拉排行榜", "/金币排行", "/金币排行榜", "/大富翁榜", "/richlist"]
 storage = false
 network_profiles = []
 service_profiles = ["coin-ranking"]

extensions/jwebot/coin-ranking/src/coin_ranking.py

@@ -7,7 +7,7 @@ PROFILE = 'coin-ranking'
 AUTHOR = 'jwebot'
 DESCRIPTION = '富榜'
 HELP_ENTRIES = ('/大富翁榜',)
-COMMANDS = ('/金币排行', '/金币排行榜', '/大富翁榜', '/richlist')
+COMMANDS = ('/摩拉排行', '/摩拉排行榜', '/金币排行', '/金币排行榜', '/大富翁榜', '/richlist')
 
 
 def _handler(context: SingleGroupContext, command: str):

extensions/jwebot/coin-ranking/tests/test_plugin.py

@@ -21,8 +21,8 @@ class PluginTests(unittest.TestCase):
         self.assertEqual(AUTHOR, 'jwebot')
         self.assertEqual(DESCRIPTION, '富榜')
         self.assertEqual(HELP_ENTRIES, ('/大富翁榜',))
-        self.assertIn('/金币排行', registered)
+        self.assertIn('/摩拉排行', registered)
-        self.assertEqual(len(registered), 4)
+        self.assertEqual(len(registered), 6)
 
 
 if __name__ == "__main__":

extensions/jwebot/coin-taffy/README.md

@@ -0,0 +1,19 @@
+# coin-taffy
+
+Author: `jwebot`
+
+Taffy摩拉
+
+Help:
+
+- `/Taffy摩拉（兼容 /Taffy金币）`
+
+Registered commands:
+
+- `/Taffy摩拉`
+- `/Taffy金币`
+- `/taffycoins`
+
+The adapter has no direct Ledger, GUI, environment, shell, or network access.
+It delegates to the reviewed `coin-taffy` Kernel service profile
+and returns a structured `ReplyPlan`.

extensions/jwebot/coin-taffy/plugin.toml

@@ -0,0 +1,14 @@
+id = "coin-taffy"
+author = "jwebot"
+description = "Taffy摩拉"
+help = ["/Taffy摩拉（兼容 /Taffy金币）"]
+version = "1.0.0"
+api_version = "1"
+entrypoint = "coin_taffy:plugin"
+type = "command"
+
+[capabilities]
+commands = ["/Taffy摩拉", "/Taffy金币", "/taffycoins"]
+storage = false
+network_profiles = []
+service_profiles = ["coin-taffy"]

extensions/jwebot/coin-taffy/src/coin_taffy.py

@@ -0,0 +1,45 @@
+"""Generated reviewed adapter for coin-taffy."""
+
+from jwebot_sdk import Event, ExtensionResult, ReplyPlan, SingleGroupContext
+
+
+PROFILE = 'coin-taffy'
+AUTHOR = 'jwebot'
+DESCRIPTION = 'Taffy摩拉'
+HELP_ENTRIES = ('/Taffy摩拉（兼容 /Taffy金币）',)
+COMMANDS = ('/Taffy摩拉', '/Taffy金币', '/taffycoins')
+
+
+def _handler(context: SingleGroupContext, command: str):
+    def handle(event: Event, arguments: str) -> ExtensionResult:
+        response = context.services.call(
+            PROFILE,
+            operation="execute",
+            payload={
+                "command": command,
+                "arguments": arguments,
+                "event": event.to_mapping(),
+            },
+        )
+        texts = response.get("texts")
+        if isinstance(texts, list) and texts and all(
+            isinstance(item, str) and item.strip() for item in texts
+        ):
+            return ExtensionResult(
+                reply=ReplyPlan.multi_text(texts),
+                metadata={"local_extension": PROFILE, "author": AUTHOR},
+            )
+        text = response.get("text")
+        if not isinstance(text, str) or not text.strip():
+            raise ValueError("local service returned invalid text")
+        return ExtensionResult(
+            reply=ReplyPlan.text(text),
+            metadata={"local_extension": PROFILE, "author": AUTHOR},
+        )
+
+    return handle
+
+
+def plugin(context: SingleGroupContext) -> None:
+    for command in COMMANDS:
+        context.commands.register(command, _handler(context, command))

extensions/jwebot/coin-taffy/tests/test_plugin.py

@@ -0,0 +1,29 @@
+import sys
+import unittest
+from pathlib import Path
+from unittest import mock
+
+
+SOURCE = Path(__file__).resolve().parents[1] / "src"
+sys.path.insert(0, str(SOURCE))
+
+from coin_taffy import AUTHOR, DESCRIPTION, HELP_ENTRIES, plugin
+
+
+class PluginTests(unittest.TestCase):
+    def test_registers_declared_commands_and_author(self):
+        context = mock.Mock()
+        plugin(context)
+        registered = [
+            call.args[0]
+            for call in context.commands.register.call_args_list
+        ]
+        self.assertEqual(AUTHOR, 'jwebot')
+        self.assertEqual(DESCRIPTION, 'Taffy摩拉')
+        self.assertEqual(HELP_ENTRIES, ('/Taffy摩拉（兼容 /Taffy金币）',))
+        self.assertIn('/Taffy摩拉', registered)
+        self.assertEqual(len(registered), 3)
+
+
+if __name__ == "__main__":
+    unittest.main()

extensions/jwebot/coin-transfer/README.md

@@ -2,7 +2,7 @@
 
 Author: `jwebot`
 
-金币转账
+摩拉转账
 
 Help:
 

extensions/jwebot/coin-transfer/plugin.toml

@@ -1,6 +1,6 @@
 id = "coin-transfer"
 author = "jwebot"
-description = "金币转账"
+description = "摩拉转账"
 help = ["/转账 对方 金额；/transfer"]
 version = "1.0.0"
 api_version = "1"

extensions/jwebot/coin-transfer/src/coin_transfer.py

@@ -5,7 +5,7 @@ from jwebot_sdk import Event, ExtensionResult, ReplyPlan, SingleGroupContext
 
 PROFILE = 'coin-transfer'
 AUTHOR = 'jwebot'
-DESCRIPTION = '金币转账'
+DESCRIPTION = '摩拉转账'
 HELP_ENTRIES = ('/转账 对方 金额；/transfer',)
 COMMANDS = ('/转账', '/transfer')
 

extensions/jwebot/coin-transfer/tests/test_plugin.py

@@ -19,7 +19,7 @@ class PluginTests(unittest.TestCase):
             for call in context.commands.register.call_args_list
         ]
         self.assertEqual(AUTHOR, 'jwebot')
-        self.assertEqual(DESCRIPTION, '金币转账')
+        self.assertEqual(DESCRIPTION, '摩拉转账')
         self.assertEqual(HELP_ENTRIES, ('/转账 对方 金额；/transfer',))
         self.assertIn('/转账', registered)
         self.assertEqual(len(registered), 2)

extensions/jwebot/single-player-games/README.md

@@ -2,19 +2,19 @@
 
 Author: `jwebot`
 
-金币游戏
+摩拉游戏
 
 Help:
 
-- `/金币 /wallet；/游戏记录 /gamehistory`
+- `/摩拉（兼容 /金币）/wallet；/游戏记录 /gamehistory`
 - `/抽奖 [金额]；/draw`
 - `/彩票 开始|报名|状态|开奖|取消；/lottery`
-- `/21点 开始 [1～500]；/blackjack start`
+- `/21点 单人开始|要牌|停牌|状态；多人启动|加入|开始|放200`
-- `/21点 要牌|停牌|状态|结束；/blackjack`
 - `/游戏帮助；/games`
 
 Registered commands:
 
+- `/摩拉`
 - `/金币`
 - `/wallet`
 - `/抽奖`

extensions/jwebot/single-player-games/plugin.toml

@@ -1,14 +1,14 @@
 id = "single-player-games"
 author = "jwebot"
-description = "金币游戏"
+description = "摩拉游戏"
-help = ["/金币 /wallet；/游戏记录 /gamehistory", "/抽奖 [金额]；/draw", "/彩票 开始|报名|状态|开奖|取消；/lottery", "/21点 开始 [1～500]；/blackjack start", "/21点 要牌|停牌|状态|结束；/blackjack", "/游戏帮助；/games"]
+help = ["/摩拉（兼容 /金币）/wallet；/游戏记录 /gamehistory", "/抽奖 [金额]；/draw", "/彩票 开始|报名|状态|开奖|取消；/lottery", "/21点 单人开始|要牌|停牌|状态；多人启动|加入|开始|放200", "/游戏帮助；/games"]
 version = "1.0.0"
 api_version = "1"
 entrypoint = "single_player_games:plugin"
 type = "command"
 
 [capabilities]
-commands = ["/金币", "/wallet", "/抽奖", "/draw", "/彩票", "/买彩票", "/lottery", "/21点", "/二十一点", "/blackjack", "/游戏记录", "/个人记录", "/gamehistory", "/游戏帮助", "/games"]
+commands = ["/摩拉", "/金币", "/wallet", "/抽奖", "/draw", "/彩票", "/买彩票", "/lottery", "/21点", "/二十一点", "/blackjack", "/游戏记录", "/个人记录", "/gamehistory", "/游戏帮助", "/games"]
 storage = false
 network_profiles = []
 service_profiles = ["single-player-games"]

extensions/jwebot/single-player-games/src/single_player_games.py

@@ -5,9 +5,9 @@ from jwebot_sdk import Event, ExtensionResult, ReplyPlan, SingleGroupContext
 
 PROFILE = 'single-player-games'
 AUTHOR = 'jwebot'
-DESCRIPTION = '金币游戏'
+DESCRIPTION = '摩拉游戏'
-HELP_ENTRIES = ('/金币 /wallet；/游戏记录 /gamehistory', '/抽奖 [金额]；/draw', '/彩票 开始|报名|状态|开奖|取消；/lottery', '/21点 开始 [1～500]；/blackjack start', '/21点 要牌|停牌|状态|结束；/blackjack', '/游戏帮助；/games')
+HELP_ENTRIES = ('/摩拉（兼容 /金币）/wallet；/游戏记录 /gamehistory', '/抽奖 [金额]；/draw', '/彩票 开始|报名|状态|开奖|取消；/lottery', '/21点 单人开始|要牌|停牌|状态；多人启动|加入|开始|放200', '/游戏帮助；/games')
-COMMANDS = ('/金币', '/wallet', '/抽奖', '/draw', '/彩票', '/买彩票', '/lottery', '/21点', '/二十一点', '/blackjack', '/游戏记录', '/个人记录', '/gamehistory', '/游戏帮助', '/games')
+COMMANDS = ('/摩拉', '/金币', '/wallet', '/抽奖', '/draw', '/彩票', '/买彩票', '/lottery', '/21点', '/二十一点', '/blackjack', '/游戏记录', '/个人记录', '/gamehistory', '/游戏帮助', '/games')
 
 
 def _handler(context: SingleGroupContext, command: str):

extensions/jwebot/single-player-games/tests/test_plugin.py

@@ -19,10 +19,10 @@ class PluginTests(unittest.TestCase):
             for call in context.commands.register.call_args_list
         ]
         self.assertEqual(AUTHOR, 'jwebot')
-        self.assertEqual(DESCRIPTION, '金币游戏')
+        self.assertEqual(DESCRIPTION, '摩拉游戏')
-        self.assertEqual(HELP_ENTRIES, ('/金币 /wallet；/游戏记录 /gamehistory', '/抽奖 [金额]；/draw', '/彩票 开始|报名|状态|开奖|取消；/lottery', '/21点 开始 [1～500]；/blackjack start', '/21点 要牌|停牌|状态|结束；/blackjack', '/游戏帮助；/games'))
+        self.assertEqual(HELP_ENTRIES, ('/摩拉（兼容 /金币）/wallet；/游戏记录 /gamehistory', '/抽奖 [金额]；/draw', '/彩票 开始|报名|状态|开奖|取消；/lottery', '/21点 单人开始|要牌|停牌|状态；多人启动|加入|开始|放200', '/游戏帮助；/games'))
-        self.assertIn('/金币', registered)
+        self.assertIn('/摩拉', registered)
-        self.assertEqual(len(registered), 15)
+        self.assertEqual(len(registered), 16)
 
 
 if __name__ == "__main__":

extensions/official/context-summary/README.md

@@ -2,7 +2,7 @@
 
 Author: `official`
 
-提取当前群最近100、200或300条本地文字的上下文
+上下文总结
 
 Help:
 

extensions/official/context-summary/plugin.toml

@@ -1,6 +1,6 @@
 id = "context-summary"
 author = "official"
-description = "提取当前群最近100、200或300条本地文字的上下文"
+description = "上下文总结"
 help = ["/上下文 [100|200|300]；/context [100|200|300]"]
 version = "1.0.0"
 api_version = "1"

extensions/official/context-summary/src/context_summary.py

@@ -5,7 +5,7 @@ from jwebot_sdk import Event, ExtensionResult, ReplyPlan, SingleGroupContext
 
 PROFILE = 'context-summary'
 AUTHOR = 'official'
-DESCRIPTION = '提取当前群最近100、200或300条本地文字的上下文'
+DESCRIPTION = '上下文总结'
 HELP_ENTRIES = ('/上下文 [100|200|300]；/context [100|200|300]',)
 COMMANDS = ('/上下文', '/context')
 

extensions/official/context-summary/tests/test_plugin.py

@@ -19,7 +19,7 @@ class PluginTests(unittest.TestCase):
             for call in context.commands.register.call_args_list
         ]
         self.assertEqual(AUTHOR, 'official')
-        self.assertEqual(DESCRIPTION, '提取当前群最近100、200或300条本地文字的上下文')
+        self.assertEqual(DESCRIPTION, '上下文总结')
         self.assertEqual(HELP_ENTRIES, ('/上下文 [100|200|300]；/context [100|200|300]',))
         self.assertIn('/上下文', registered)
         self.assertEqual(len(registered), 2)

jwebot_release_update.py

@@ -0,0 +1,599 @@
+#!/usr/bin/env python3
+"""Apply the mandatory post-release Mora grant and group announcement."""
+
+from __future__ import annotations
+
+import argparse
+import hashlib
+import json
+import os
+import re
+import sqlite3
+import subprocess
+import sys
+from collections.abc import Callable
+from datetime import datetime, timezone
+from pathlib import Path
+from typing import Any
+
+
+ROOT = Path(__file__).resolve().parent
+sys.path[:0] = [
+    str(ROOT / "packages" / "jwebot-official-extensions" / "src"),
+    str(ROOT / "packages" / "jwebot-sdk" / "src"),
+]
+
+from jwebot_official_extensions import __version__ as OFFICIAL_VERSION  # noqa: E402
+from jwebot_official_extensions.player_games import (  # noqa: E402
+    append_coin_event,
+    normalize_player_account,
+)
+
+
+COMPENSATION_AMOUNT = 200
+MAX_BALANCE = 1_000_000_000
+MAX_SUMMARY_CHARS = 300
+MAX_ANNOUNCEMENT_CHARS = 900
+RELEASE_PATTERN = re.compile(
+    r"^beta (?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)\.(?P<date>\d{4})$"
+)
+CHANGELOG_HEADING = re.compile(
+    r"^##\s+(?P<release>beta \d+\.\d+\.\d+\.\d{4})"
+    r"(?:\s+-\s+\d{4}-\d{2}-\d{2})?\s*$"
+)
+
+
+class ReleaseUpdateError(RuntimeError):
+    """Raised when release settlement cannot complete safely."""
+
+
+class AnnouncementUncertainError(ReleaseUpdateError):
+    """Raised when an announcement may have been sent without a final receipt."""
+
+
+def utc_now() -> datetime:
+    return datetime.now(timezone.utc)
+
+
+def isoformat_utc(value: datetime) -> str:
+    return value.astimezone(timezone.utc).isoformat(
+        timespec="milliseconds"
+    ).replace("+00:00", "Z")
+
+
+def release_to_pep440(release: str) -> str:
+    match = RELEASE_PATTERN.fullmatch(release)
+    if match is None:
+        raise ReleaseUpdateError(f"unsupported release format: {release}")
+    return (
+        f"{match.group('major')}.{match.group('minor')}."
+        f"{match.group('patch')}b1+{match.group('date')}"
+    )
+
+
+def read_release(path: Path) -> str:
+    try:
+        release = path.read_text(encoding="utf-8").strip()
+    except OSError as error:
+        raise ReleaseUpdateError("cannot read VERSION") from error
+    release_to_pep440(release)
+    return release
+
+
+def read_changelog_summary(path: Path, release: str) -> str:
+    try:
+        lines = path.read_text(encoding="utf-8").splitlines()
+    except OSError as error:
+        raise ReleaseUpdateError("cannot read CHANGELOG.md") from error
+    section_index: int | None = None
+    for index, line in enumerate(lines):
+        match = CHANGELOG_HEADING.fullmatch(line)
+        if match is not None and match.group("release") == release:
+            section_index = index
+            break
+    if section_index is None:
+        raise ReleaseUpdateError("release is missing from CHANGELOG.md")
+    parts: list[str] = []
+    collecting = False
+    for line in lines[section_index + 1 :]:
+        if line.startswith("## "):
+            break
+        if line.startswith("- "):
+            if collecting:
+                break
+            collecting = True
+            parts.append(line[2:].strip())
+            continue
+        if collecting:
+            if line.startswith("  "):
+                parts.append(line.strip())
+                continue
+            if not line.strip():
+                break
+    summary = " ".join(" ".join(parts).split())
+    if not summary:
+        raise ReleaseUpdateError("release changelog needs a first summary bullet")
+    if len(summary) > MAX_SUMMARY_CHARS:
+        raise ReleaseUpdateError("release changelog summary is too long")
+    return summary
+
+
+def build_announcement(release: str, summary: str) -> str:
+    announcement = (
+        "亲爱的群友：\n"
+        f"{release} 更新现已完成。\n"
+        f"更新内容：{summary}\n"
+        f"更新补偿：摩拉 x{COMPENSATION_AMOUNT}\n"
+        "补偿范围：截至本公告发布前已创建摩拉账户的玩家\n"
+        "发放方式：补偿已直接到账，无需领取\n"
+        "感谢大家对 Taffy 的支持与理解，祝大家游戏愉快！"
+    )
+    if len(announcement) > MAX_ANNOUNCEMENT_CHARS:
+        raise ReleaseUpdateError("release announcement is too long")
+    return announcement
+
+
+def compensation_key(release: str) -> str:
+    return f"release_compensation:{release}"
+
+
+def announcement_key(release: str, group_fingerprint: str) -> str:
+    group_hash = hashlib.sha256(group_fingerprint.encode()).hexdigest()
+    return f"release_announcement:{release}:{group_hash}"
+
+
+def release_slug(release: str) -> str:
+    return re.sub(r"[^a-zA-Z0-9]+", "-", release).strip("-").lower()
+
+
+def checked_marker(raw: str, release: str) -> dict[str, Any]:
+    try:
+        marker = json.loads(raw)
+    except json.JSONDecodeError as error:
+        raise ReleaseUpdateError("release compensation marker is invalid") from error
+    if (
+        not isinstance(marker, dict)
+        or marker.get("release") != release
+        or marker.get("amount") != COMPENSATION_AMOUNT
+        or not isinstance(marker.get("group_fingerprints"), list)
+    ):
+        raise ReleaseUpdateError("release compensation marker does not match")
+    return marker
+
+
+def backup_database(
+    source: sqlite3.Connection,
+    backup_dir: Path,
+    release: str,
+    now: datetime,
+) -> Path:
+    backup_dir.mkdir(mode=0o700, parents=True, exist_ok=True)
+    os.chmod(backup_dir, 0o700)
+    stamp = now.astimezone(timezone.utc).strftime("%Y%m%dT%H%M%S%fZ")
+    destination = backup_dir / (
+        f"edge-before-{release_slug(release)}-compensation-{stamp}.sqlite3"
+    )
+    target = sqlite3.connect(destination)
+    try:
+        source.backup(target)
+        if target.execute("PRAGMA quick_check").fetchone()[0] != "ok":
+            raise ReleaseUpdateError("release compensation backup failed integrity check")
+    finally:
+        target.close()
+    destination.chmod(0o600)
+    return destination
+
+
+def load_compensation_marker(
+    connection: sqlite3.Connection,
+    release: str,
+) -> dict[str, Any] | None:
+    row = connection.execute(
+        "SELECT value FROM metadata WHERE key = ?",
+        (compensation_key(release),),
+    ).fetchone()
+    if row is None:
+        return None
+    return checked_marker(str(row[0]), release)
+
+
+def apply_compensation(
+    database_path: Path,
+    backup_dir: Path,
+    release: str,
+    *,
+    now: datetime | None = None,
+) -> tuple[dict[str, Any], str]:
+    applied_at = now or utc_now()
+    connection = sqlite3.connect(database_path, timeout=30)
+    connection.row_factory = sqlite3.Row
+    try:
+        if connection.execute("PRAGMA quick_check").fetchone()[0] != "ok":
+            raise ReleaseUpdateError("Edge SQLite integrity check failed")
+        existing = load_compensation_marker(connection, release)
+        if existing is not None:
+            return existing, "already_applied"
+        backup_path = backup_database(
+            connection,
+            backup_dir,
+            release,
+            applied_at,
+        )
+        connection.execute("BEGIN IMMEDIATE")
+        try:
+            existing = load_compensation_marker(connection, release)
+            if existing is not None:
+                connection.execute("COMMIT")
+                return existing, "already_applied"
+            rows = connection.execute(
+                """
+                SELECT group_fingerprint, person_key, state_json
+                FROM player_accounts
+                ORDER BY group_fingerprint, person_key
+                """
+            ).fetchall()
+            groups = sorted({str(row["group_fingerprint"]) for row in rows})
+            old_balance_sum = 0
+            timestamp = isoformat_utc(applied_at)
+            event_reason = f"{release} 版本更新补偿"
+            for row in rows:
+                try:
+                    state = json.loads(str(row["state_json"]))
+                except json.JSONDecodeError as error:
+                    raise ReleaseUpdateError("Mora account state is invalid") from error
+                if not isinstance(state, dict):
+                    raise ReleaseUpdateError("Mora account state is invalid")
+                normalize_player_account(state)
+                balance = state.get("balance")
+                if (
+                    type(balance) is not int
+                    or not 0 <= balance <= MAX_BALANCE - COMPENSATION_AMOUNT
+                ):
+                    raise ReleaseUpdateError("Mora account balance is invalid")
+                old_balance_sum += balance
+                state["balance"] = balance + COMPENSATION_AMOUNT
+                append_coin_event(state, COMPENSATION_AMOUNT, event_reason)
+                state["coin_events"][-1]["created_at"] = applied_at.timestamp()
+                connection.execute(
+                    """
+                    UPDATE player_accounts
+                    SET state_json = ?, updated_at = ?
+                    WHERE group_fingerprint = ? AND person_key = ?
+                    """,
+                    (
+                        json.dumps(state, ensure_ascii=False),
+                        timestamp,
+                        row["group_fingerprint"],
+                        row["person_key"],
+                    ),
+                )
+            marker = {
+                "release": release,
+                "amount": COMPENSATION_AMOUNT,
+                "accounts": len(rows),
+                "group_fingerprints": groups,
+                "old_balance_sum": old_balance_sum,
+                "new_balance_sum": (
+                    old_balance_sum + COMPENSATION_AMOUNT * len(rows)
+                ),
+                "applied_at": timestamp,
+                "backup": backup_path.name,
+            }
+            connection.execute(
+                """
+                INSERT INTO metadata (key, value, updated_at)
+                VALUES (?, ?, ?)
+                """,
+                (
+                    compensation_key(release),
+                    json.dumps(marker, ensure_ascii=False, separators=(",", ":")),
+                    timestamp,
+                ),
+            )
+            connection.execute("COMMIT")
+            return marker, "applied"
+        except Exception:
+            if connection.in_transaction:
+                connection.execute("ROLLBACK")
+            raise
+    finally:
+        connection.close()
+
+
+def metadata_value(
+    connection: sqlite3.Connection,
+    key: str,
+) -> dict[str, Any] | None:
+    row = connection.execute(
+        "SELECT value FROM metadata WHERE key = ?",
+        (key,),
+    ).fetchone()
+    if row is None:
+        return None
+    try:
+        value = json.loads(str(row[0]))
+    except json.JSONDecodeError as error:
+        raise ReleaseUpdateError("announcement marker is invalid") from error
+    if not isinstance(value, dict):
+        raise ReleaseUpdateError("announcement marker is invalid")
+    return value
+
+
+def store_metadata(
+    connection: sqlite3.Connection,
+    key: str,
+    value: dict[str, Any],
+    updated_at: str,
+) -> None:
+    connection.execute(
+        """
+        INSERT INTO metadata (key, value, updated_at)
+        VALUES (?, ?, ?)
+        ON CONFLICT(key) DO UPDATE
+        SET value = excluded.value, updated_at = excluded.updated_at
+        """,
+        (
+            key,
+            json.dumps(value, ensure_ascii=False, separators=(",", ":")),
+            updated_at,
+        ),
+    )
+
+
+def run_delivery_helper(
+    delivery_helper: Path,
+    group_fingerprint: str,
+    announcement: str,
+    event_id: str,
+) -> None:
+    payload = {
+        "group_fingerprint": group_fingerprint,
+        "reply": announcement,
+        "event_id": event_id,
+    }
+    try:
+        result = subprocess.run(
+            [sys.executable, str(delivery_helper)],
+            input=json.dumps(payload, ensure_ascii=False),
+            text=True,
+            capture_output=True,
+            timeout=70,
+            check=False,
+        )
+    except (OSError, subprocess.TimeoutExpired) as error:
+        raise AnnouncementUncertainError("announcement transport failed") from error
+    if result.returncode != 0:
+        raise AnnouncementUncertainError(
+            f"announcement helper failed with return code {result.returncode}"
+        )
+
+
+def deliver_announcements(
+    database_path: Path,
+    release: str,
+    groups: list[str],
+    announcement: str,
+    delivery_helper: Path,
+    *,
+    retry_uncertain: bool = False,
+    deliverer: Callable[[Path, str, str, str], None] = run_delivery_helper,
+    now: Callable[[], datetime] = utc_now,
+) -> dict[str, int]:
+    delivered = 0
+    skipped = 0
+    for group in groups:
+        key = announcement_key(release, group)
+        group_hash = hashlib.sha256(group.encode()).hexdigest()[:12]
+        connection = sqlite3.connect(database_path, timeout=30)
+        try:
+            existing = metadata_value(connection, key)
+            state = str(existing.get("state", "")) if existing else ""
+            if state == "delivered":
+                skipped += 1
+                continue
+            if state in {"sending", "uncertain"} and not retry_uncertain:
+                raise AnnouncementUncertainError(
+                    "previous announcement delivery is uncertain; "
+                    "visually confirm before using --retry-uncertain-announcement"
+                )
+            attempt = int(existing.get("attempt", 0)) + 1 if existing else 1
+            started_at = isoformat_utc(now())
+            with connection:
+                store_metadata(
+                    connection,
+                    key,
+                    {
+                        "release": release,
+                        "state": "sending",
+                        "attempt": attempt,
+                        "group_hash": group_hash,
+                        "started_at": started_at,
+                    },
+                    started_at,
+                )
+        finally:
+            connection.close()
+
+        event_id = f"release-{release_slug(release)}-{group_hash}"
+        try:
+            deliverer(
+                delivery_helper,
+                group,
+                announcement,
+                event_id,
+            )
+        except Exception as error:
+            failed_at = isoformat_utc(now())
+            connection = sqlite3.connect(database_path, timeout=30)
+            try:
+                with connection:
+                    store_metadata(
+                        connection,
+                        key,
+                        {
+                            "release": release,
+                            "state": "uncertain",
+                            "attempt": attempt,
+                            "group_hash": group_hash,
+                            "started_at": started_at,
+                            "failed_at": failed_at,
+                            "error_type": type(error).__name__[:80],
+                        },
+                        failed_at,
+                    )
+            finally:
+                connection.close()
+            if isinstance(error, ReleaseUpdateError):
+                raise
+            raise AnnouncementUncertainError(
+                "announcement delivery is uncertain"
+            ) from error
+
+        completed_at = isoformat_utc(now())
+        connection = sqlite3.connect(database_path, timeout=30)
+        try:
+            with connection:
+                store_metadata(
+                    connection,
+                    key,
+                    {
+                        "release": release,
+                        "state": "delivered",
+                        "attempt": attempt,
+                        "group_hash": group_hash,
+                        "started_at": started_at,
+                        "delivered_at": completed_at,
+                    },
+                    completed_at,
+                )
+        finally:
+            connection.close()
+        delivered += 1
+    return {"delivered": delivered, "skipped": skipped}
+
+
+def run_release_update(
+    *,
+    release_file: Path,
+    changelog: Path,
+    database_path: Path,
+    backup_dir: Path,
+    delivery_helper: Path,
+    retry_uncertain: bool = False,
+    deliverer: Callable[[Path, str, str, str], None] = run_delivery_helper,
+    applied_at: datetime | None = None,
+    clock: Callable[[], datetime] = utc_now,
+) -> dict[str, Any]:
+    release = read_release(release_file)
+    expected_package_version = release_to_pep440(release)
+    if OFFICIAL_VERSION != expected_package_version:
+        raise ReleaseUpdateError(
+            "installed package version does not match VERSION"
+        )
+    summary = read_changelog_summary(changelog, release)
+    announcement = build_announcement(release, summary)
+    marker, compensation_state = apply_compensation(
+        database_path,
+        backup_dir,
+        release,
+        now=applied_at,
+    )
+    groups = [str(value) for value in marker["group_fingerprints"]]
+    announcement_result = deliver_announcements(
+        database_path,
+        release,
+        groups,
+        announcement,
+        delivery_helper,
+        retry_uncertain=retry_uncertain,
+        deliverer=deliverer,
+        now=clock,
+    )
+    connection = sqlite3.connect(
+        f"file:{database_path}?mode=ro",
+        uri=True,
+    )
+    try:
+        if connection.execute("PRAGMA quick_check").fetchone()[0] != "ok":
+            raise ReleaseUpdateError("post-release SQLite integrity check failed")
+    finally:
+        connection.close()
+    return {
+        "status": "ok",
+        "release": release,
+        "compensation": compensation_state,
+        "amount_each": COMPENSATION_AMOUNT,
+        "accounts": int(marker["accounts"]),
+        "groups": len(groups),
+        "announcements_delivered": announcement_result["delivered"],
+        "announcements_skipped": announcement_result["skipped"],
+        "backup": str(marker["backup"]),
+    }
+
+
+def parse_args() -> argparse.Namespace:
+    home = Path.home()
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument(
+        "--release-file",
+        type=Path,
+        default=ROOT / "VERSION",
+    )
+    parser.add_argument(
+        "--changelog",
+        type=Path,
+        default=ROOT / "CHANGELOG.md",
+    )
+    parser.add_argument(
+        "--database",
+        type=Path,
+        default=home / ".local/state/jwebot-edge/edge.sqlite3",
+    )
+    parser.add_argument(
+        "--backup-dir",
+        type=Path,
+        default=home / ".local/state/jwebot/admin-backups",
+    )
+    parser.add_argument(
+        "--delivery-helper",
+        type=Path,
+        default=ROOT / "jwebot_edge_delivery.py",
+    )
+    parser.add_argument(
+        "--retry-uncertain-announcement",
+        action="store_true",
+        help="retry only after visually confirming that the prior send did not land",
+    )
+    return parser.parse_args()
+
+
+def main() -> int:
+    args = parse_args()
+    try:
+        result = run_release_update(
+            release_file=args.release_file,
+            changelog=args.changelog,
+            database_path=args.database,
+            backup_dir=args.backup_dir,
+            delivery_helper=args.delivery_helper,
+            retry_uncertain=args.retry_uncertain_announcement,
+        )
+    except ReleaseUpdateError as error:
+        print(
+            json.dumps(
+                {
+                    "status": "error",
+                    "error_type": type(error).__name__,
+                    "message": str(error),
+                },
+                ensure_ascii=False,
+            ),
+            file=sys.stderr,
+        )
+        return 1
+    print(json.dumps(result, ensure_ascii=False, sort_keys=True))
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

packages/jwebot-config/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "jwebot-config"
-version = "1.0.0b1+1016"
+version = "1.0.4b1+1016"
 description = "Profile-based configuration compiler for JWeBot"
 requires-python = ">=3.9"
 dependencies = [

packages/jwebot-config/src/jwebot_config/__init__.py

@@ -36,4 +36,4 @@ __all__ = [
     "write_default_profile",
 ]
 
-__version__ = "1.0.0b1+1016"
+__version__ = "1.0.4b1+1016"

packages/jwebot-extension-host/pyproject.toml

@@ -4,12 +4,12 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "jwebot-extension-host"
-version = "1.0.0b1+1016"
+version = "1.0.4b1+1016"
 description = "Single-release, single-invocation process host for JWeBot extensions"
 requires-python = ">=3.9"
 dependencies = [
-  "jwebot-sdk==1.0.0b1+1016",
+  "jwebot-sdk==1.0.4b1+1016",
-  "jwebot-kernel==1.0.0b1+1016",
+  "jwebot-kernel==1.0.4b1+1016",
 ]
 
 [project.scripts]

packages/jwebot-extension-host/src/jwebot_extension_host/__init__.py

@@ -16,4 +16,4 @@ __all__ = [
     "process_request",
 ]
 
-__version__ = "1.0.0b1+1016"
+__version__ = "1.0.4b1+1016"

packages/jwebot-kernel/pyproject.toml

@@ -4,13 +4,13 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "jwebot-kernel"
-version = "1.0.0b1+1016"
+version = "1.0.4b1+1016"
 description = "Stable routing and extension authority for JWeBot"
 requires-python = ">=3.9"
 dependencies = [
-  "jwebot-config==1.0.0b1+1016",
+  "jwebot-config==1.0.4b1+1016",
-  "jwebot-sdk==1.0.0b1+1016",
+  "jwebot-sdk==1.0.4b1+1016",
-  "jwebot-official-extensions==1.0.0b1+1016",
+  "jwebot-official-extensions==1.0.4b1+1016",
   "tomli>=2; python_version < '3.11'",
 ]
 

packages/jwebot-kernel/src/jwebot_kernel/__init__.py

@@ -107,4 +107,4 @@ __all__ = [
     "verify_installed_release_signature",
 ]
 
-__version__ = "1.0.0b1+1016"
+__version__ = "1.0.4b1+1016"

packages/jwebot-official-extensions/pyproject.toml

@@ -4,11 +4,11 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "jwebot-official-extensions"
-version = "1.0.0b1+1016"
+version = "1.0.4b1+1016"
 description = "Reviewed local command adapters and broker implementation for JWeBot"
 requires-python = ">=3.9"
 dependencies = [
-  "jwebot-sdk==1.0.0b1+1016",
+  "jwebot-sdk==1.0.4b1+1016",
 ]
 
 [tool.setuptools.packages.find]

packages/jwebot-official-extensions/src/jwebot_official_extensions/__init__.py

@@ -90,4 +90,4 @@ __all__ = [
     "render_feature_list",
 ]
 
-__version__ = "1.0.0b1+1016"
+__version__ = "1.0.4b1+1016"

packages/jwebot-official-extensions/src/jwebot_official_extensions/adapters/coin_ranking.py

@@ -7,7 +7,7 @@ PROFILE = 'coin-ranking'
 AUTHOR = 'jwebot'
 DESCRIPTION = '富榜'
 HELP_ENTRIES = ('/大富翁榜',)
-COMMANDS = ('/金币排行', '/金币排行榜', '/大富翁榜', '/richlist')
+COMMANDS = ('/摩拉排行', '/摩拉排行榜', '/金币排行', '/金币排行榜', '/大富翁榜', '/richlist')
 
 
 def _handler(context: SingleGroupContext, command: str):

packages/jwebot-official-extensions/src/jwebot_official_extensions/adapters/coin_taffy.py

@@ -0,0 +1,45 @@
+"""Generated reviewed adapter for coin-taffy."""
+
+from jwebot_sdk import Event, ExtensionResult, ReplyPlan, SingleGroupContext
+
+
+PROFILE = 'coin-taffy'
+AUTHOR = 'jwebot'
+DESCRIPTION = 'Taffy摩拉'
+HELP_ENTRIES = ('/Taffy摩拉（兼容 /Taffy金币）',)
+COMMANDS = ('/Taffy摩拉', '/Taffy金币', '/taffycoins')
+
+
+def _handler(context: SingleGroupContext, command: str):
+    def handle(event: Event, arguments: str) -> ExtensionResult:
+        response = context.services.call(
+            PROFILE,
+            operation="execute",
+            payload={
+                "command": command,
+                "arguments": arguments,
+                "event": event.to_mapping(),
+            },
+        )
+        texts = response.get("texts")
+        if isinstance(texts, list) and texts and all(
+            isinstance(item, str) and item.strip() for item in texts
+        ):
+            return ExtensionResult(
+                reply=ReplyPlan.multi_text(texts),
+                metadata={"local_extension": PROFILE, "author": AUTHOR},
+            )
+        text = response.get("text")
+        if not isinstance(text, str) or not text.strip():
+            raise ValueError("local service returned invalid text")
+        return ExtensionResult(
+            reply=ReplyPlan.text(text),
+            metadata={"local_extension": PROFILE, "author": AUTHOR},
+        )
+
+    return handle
+
+
+def plugin(context: SingleGroupContext) -> None:
+    for command in COMMANDS:
+        context.commands.register(command, _handler(context, command))

packages/jwebot-official-extensions/src/jwebot_official_extensions/adapters/coin_transfer.py

@@ -5,7 +5,7 @@ from jwebot_sdk import Event, ExtensionResult, ReplyPlan, SingleGroupContext
 
 PROFILE = 'coin-transfer'
 AUTHOR = 'jwebot'
-DESCRIPTION = '金币转账'
+DESCRIPTION = '摩拉转账'
 HELP_ENTRIES = ('/转账 对方 金额；/transfer',)
 COMMANDS = ('/转账', '/transfer')
 

packages/jwebot-official-extensions/src/jwebot_official_extensions/adapters/context_summary.py

@@ -5,7 +5,7 @@ from jwebot_sdk import Event, ExtensionResult, ReplyPlan, SingleGroupContext
 
 PROFILE = 'context-summary'
 AUTHOR = 'official'
-DESCRIPTION = '提取当前群最近100、200或300条本地文字的上下文'
+DESCRIPTION = '上下文总结'
 HELP_ENTRIES = ('/上下文 [100|200|300]；/context [100|200|300]',)
 COMMANDS = ('/上下文', '/context')
 

packages/jwebot-official-extensions/src/jwebot_official_extensions/adapters/single_player_games.py

@@ -5,9 +5,9 @@ from jwebot_sdk import Event, ExtensionResult, ReplyPlan, SingleGroupContext
 
 PROFILE = 'single-player-games'
 AUTHOR = 'jwebot'
-DESCRIPTION = '金币游戏'
+DESCRIPTION = '摩拉游戏'
-HELP_ENTRIES = ('/金币 /wallet；/游戏记录 /gamehistory', '/抽奖 [金额]；/draw', '/彩票 开始|报名|状态|开奖|取消；/lottery', '/21点 开始 [1～500]；/blackjack start', '/21点 要牌|停牌|状态|结束；/blackjack', '/游戏帮助；/games')
+HELP_ENTRIES = ('/摩拉（兼容 /金币）/wallet；/游戏记录 /gamehistory', '/抽奖 [金额]；/draw', '/彩票 开始|报名|状态|开奖|取消；/lottery', '/21点 单人开始|要牌|停牌|状态；多人启动|加入|开始|放200', '/游戏帮助；/games')
-COMMANDS = ('/金币', '/wallet', '/抽奖', '/draw', '/彩票', '/买彩票', '/lottery', '/21点', '/二十一点', '/blackjack', '/游戏记录', '/个人记录', '/gamehistory', '/游戏帮助', '/games')
+COMMANDS = ('/摩拉', '/金币', '/wallet', '/抽奖', '/draw', '/彩票', '/买彩票', '/lottery', '/21点', '/二十一点', '/blackjack', '/游戏记录', '/个人记录', '/gamehistory', '/游戏帮助', '/games')
 
 
 def _handler(context: SingleGroupContext, command: str):

packages/jwebot-official-extensions/src/jwebot_official_extensions/cs_autochess.py

@@ -956,7 +956,7 @@ def _shop_text(state: dict[str, Any], player: dict[str, Any]) -> str:
     lines = [
         (
             f"{_player_prefix(state, player)}第 {state['round_number']} 回合"
-            f" {phase_name}；金币 {player['gold']}；生命 {player['hp']}；"
+            f" {phase_name}；摩拉 {player['gold']}；生命 {player['hp']}；"
             f"教练 {player['coach_level']} 级；战术{player['tactic']}；"
             f"阵容 {power['power']:.1f}；商店{lock}"
         )
@@ -1143,7 +1143,7 @@ def _auto_manage_ai(
                 )
         text = (
             f"{player['display_name']} 自动运营：{'、'.join(actions)}；"
-            f"金币 {player['gold']}，五人首发战力 "
+            f"摩拉 {player['gold']}，五人首发战力 "
             f"{lineup_power(player)['power']:.1f}。"
         )
         stories.append(
@@ -1571,7 +1571,7 @@ def _settle_pve(
         )
         text = (
             f"{player['display_name']} {'完成' if won else '未完成'}"
-            f"{event_name}目标 {adjusted:.1f}/{target}，获得 {reward} 金币。"
+            f"{event_name}目标 {adjusted:.1f}/{target}，获得 {reward} 摩拉。"
         )
         stories.append(_story(state, "pve_result", text, result))
         lines.append(_host_text(text))
@@ -2326,7 +2326,7 @@ def _team_power_line(state: dict[str, Any], key: str) -> str:
         else f"{player.get('coach_level', 1)}级"
     )
     return (
-        f"{player['display_name']} #{state['season_number']}：金币 {player['gold']}；"
+        f"{player['display_name']} #{state['season_number']}：摩拉 {player['gold']}；"
         f"生命 {player['hp']}；教练 {coach_text}；战术{player['tactic']}；"
         f"阵容 {power['power']:.1f}；商店{'已锁' if player['shop_locked'] else '未锁'}"
     )
@@ -2417,7 +2417,7 @@ def _training_match_part(
                 f"{_team_name(winner)} 以 {item['score']} 拿下本图。"
             ),
             (
-                f"金币：{_team_name(winner)} +{TRAINING_WIN_GOLD}，"
+                f"摩拉：{_team_name(winner)} +{TRAINING_WIN_GOLD}，"
                 f"{_team_name(loser)} +{TRAINING_LOSS_GOLD}。"
             ),
         ]
@@ -2433,7 +2433,7 @@ def _training_bye_part(state: dict[str, Any], key: str, round_number: int) -> st
     )
     return (
         f"训练赛第 {round_number} 轮：{_team_name(player)} 与本地陪练队完成热身。\n"
-        f"{hook}\n金币：{_team_name(player)} +{TRAINING_BYE_GOLD}。"
+        f"{hook}\n摩拉：{_team_name(player)} +{TRAINING_BYE_GOLD}。"
     )
 
 
@@ -2461,7 +2461,7 @@ def _settle_training_round(
     if cursor == 0:
         intro = (
             f"欢迎来到{state['event_title']}！你们已经到达酒店宾馆，"
-            f"训练赛共 {len(rounds)} 轮。训练赛不扣生命，只产出运营金币。"
+            f"训练赛共 {len(rounds)} 轮。训练赛不扣生命，只产出运营摩拉。"
         )
         parts.append(intro)
         stories.append(_story(state, "training_open", intro))
@@ -2670,7 +2670,7 @@ def _series_report_parts(
     state["players"][winner_key]["gold"] = int(state["players"][winner_key]["gold"]) + reward
     final_text = (
         f"系列赛结束，{_team_name(state['players'][winner_key])} 以 {result['score']} 击败 "
-        f"{_team_name(state['players'][loser_key])}，获得 {reward} 金币。"
+        f"{_team_name(state['players'][loser_key])}，获得 {reward} 摩拉。"
         "这笔钱可以马上用于 /刷新、/选1～/选6 或锁店等下一轮运营。"
     )
     parts[-1] = parts[-1] + "\n" + final_text
@@ -2745,7 +2745,7 @@ def _special_event_parts(
         )
         text = (
             f"{_team_name(player)} {'完成' if won else '勉强应付'}{event_kind}："
-            f"{hook}获得 {reward} 金币。"
+            f"{hook}获得 {reward} 摩拉。"
         )
         parts.append(text)
         stories.append(_story(state, "special_event", text, {"person_key": key}))
@@ -3141,7 +3141,7 @@ class CsMajorAutochess:
             story = _story(state, "tactic_selected", text, {"person_key": person_key, "tactic": tactic})
             return respond(state, text, stories=[story])
         if action == "coach":
-            return respond(state, "教练现在由第一轮选秀决定，本届不能用金币升级教练。")
+            return respond(state, "教练现在由第一轮选秀决定，本届不能用摩拉升级教练。")
         if action == "buy":
             values = _parse_indices(arguments)
             if len(values) != 1 or not 1 <= values[0] <= SHOP_SIZE:
@@ -3152,7 +3152,7 @@ class CsMajorAutochess:
                 return respond(state, "这个商店位置已经购买或没有候选")
             card = CARD_BY_ID[str(card_id)]
             if int(player["gold"]) < int(card["cost"]):
-                return respond(state, f"金币不足：{card['name']} 需要 {card['cost']}，当前 {player['gold']}")
+                return respond(state, f"摩拉不足：{card['name']} 需要 {card['cost']}，当前 {player['gold']}")
             existing = any(
                 unit["card_id"] == card_id
                 for unit in [*player["board"], *player["bench"]]
@@ -3166,7 +3166,7 @@ class CsMajorAutochess:
             star, upgraded = _merge_purchase(player, str(card_id))
             player["shop"][slot] = None
             text = (
-                f"{display_name} 签下 {card['name']}，花费 {card['cost']} 金币；"
+                f"{display_name} 签下 {card['name']}，花费 {card['cost']} 摩拉；"
                 f"当前 {star} 星，余额 {player['gold']}。"
             )
             if upgraded:
@@ -3175,7 +3175,7 @@ class CsMajorAutochess:
             return respond(state, text, stories=[story])
         if action == "refresh":
             if int(player["gold"]) < REFRESH_COST:
-                return respond(state, f"刷新需要 {REFRESH_COST} 金币，当前 {player['gold']}")
+                return respond(state, f"刷新需要 {REFRESH_COST} 摩拉，当前 {player['gold']}")
             player["gold"] -= REFRESH_COST
             player["shop"] = _draw_shop(
                 state,
@@ -3222,7 +3222,7 @@ class CsMajorAutochess:
             refund = int(card["cost"]) * int(unit["copies"])
             player["gold"] += refund
             state["pool"][card["id"]] += int(unit["copies"])
-            return respond(state, f"{display_name} 出售 {card['name']}，返还 {refund} 金币。")
+            return respond(state, f"{display_name} 出售 {card['name']}，返还 {refund} 摩拉。")
 
         if action == "ready":
             if person_key not in state.get("ready_keys", []):

packages/jwebot-official-extensions/src/jwebot_official_extensions/data/cs_autochess/narratives.json

@@ -11,7 +11,7 @@
       "队伍如果阵容完整，就更容易把这次插曲处理成奖金。",
       "这次挑战只出现一次，处理完就回到小组赛主线。",
       "主舞台彩排占用了热身时间，教练的管理能力开始生效。",
-      "赞助商要求展示新战术，选错方向也不会淘汰，只会少拿金币。",
+      "赞助商要求展示新战术，选错方向也不会淘汰，只会少拿摩拉。",
       "现场采访提到了前一场失利，队员心态需要稳住。",
       "这类挑战不扣生命，目的只是给运营阶段增加变数。",
       "处理好这次插曲的队伍，会在下一场前多一点经济余地。"
@@ -141,21 +141,21 @@
       "{team} 这波 {count} 连分让地图气质变了，所有枪位都开始往前压。"
     ],
     "training_hooks": [
-      "训练赛不计入小组胜负，但会发金币，方便后续刷新商店。",
+      "训练赛不计入小组胜负，但会发摩拉，方便后续刷新商店。",
       "这轮训练更像实战热身，每队都会遇到同场参赛对手。",
       "训练赛重点不是淘汰谁，而是检验当前阵容能不能打完整张图。",
       "这轮结束后可以 D 牌、锁店或调首发，下一轮再检验成果。",
       "AI 战队也会参与训练，对局结果会写入档案。",
       "如果阵容缺少指挥，训练赛的中期回合会更容易失控。",
       "如果你想追羁绊，现在是最适合刷新商店的窗口。",
-      "训练赛的金币很重要，别把每次刷新都当成免费试错。",
+      "训练赛的摩拉很重要，别把每次刷新都当成免费试错。",
       "这一阶段没有生命扣除，重点是运营和阵容成型。",
       "训练赛的地图会轮换，不会一直用同一个固定比分模板。",
       "每轮训练结束后，真人都需要发送 /结束回合 才会进入下一轮。",
       "训练赛会按轮次播报，避免一次性刷出一大串结果。",
       "如果出现轮空，会安排本地陪练队补一张 BO1。",
       "训练阶段适合观察战术克制，火力、稳定、残局都会影响回合走向。",
-      "连胜会带来金币，但失败也有保底，保证测试局能继续运营。",
+      "连胜会带来摩拉，但失败也有保底，保证测试局能继续运营。",
       "这部分是纯文字自走棋的节奏层，不再是单条目标数值判定。"
     ]
   },

packages/jwebot-official-extensions/src/jwebot_official_extensions/descriptors.py

@@ -40,12 +40,18 @@ LOCAL_COMMAND_ALIASES = {
     "steam": "steam_friends",
     "上下文": "context_summary",
     "context": "context_summary",
+    "摩拉": "wallet",
     "金币": "wallet",
     "wallet": "wallet",
+    "摩拉排行": "coin_ranking",
+    "摩拉排行榜": "coin_ranking",
     "金币排行": "coin_ranking",
     "金币排行榜": "coin_ranking",
     "大富翁榜": "coin_ranking",
     "richlist": "coin_ranking",
+    "Taffy摩拉": "taffy_coins",
+    "Taffy金币": "taffy_coins",
+    "taffycoins": "taffy_coins",
     "转账": "transfer",
     "transfer": "transfer",
     "贷款": "loan",
@@ -143,20 +149,19 @@ LOCAL_EXTENSIONS = (
     LocalExtensionDescriptor(
         extension_id="context-summary",
         author=OFFICIAL_AUTHOR,
-        description="提取当前群最近100、200或300条本地文字的上下文",
+        description="上下文总结",
         help_entries=("/上下文 [100|200|300]；/context [100|200|300]",),
         command_kinds=("context_summary",),
     ),
     LocalExtensionDescriptor(
         extension_id="single-player-games",
         author=JWEBOT_AUTHOR,
-        description="金币游戏",
+        description="摩拉游戏",
         help_entries=(
-            "/金币 /wallet；/游戏记录 /gamehistory",
+            "/摩拉（兼容 /金币）/wallet；/游戏记录 /gamehistory",
             "/抽奖 [金额]；/draw",
             "/彩票 开始|报名|状态|开奖|取消；/lottery",
-            "/21点 开始 [1～500]；/blackjack start",
+            "/21点 单人开始|要牌|停牌|状态；多人启动|加入|开始|放200",
-            "/21点 要牌|停牌|状态|结束；/blackjack",
             "/游戏帮助；/games",
         ),
         command_kinds=(
@@ -171,7 +176,7 @@ LOCAL_EXTENSIONS = (
     LocalExtensionDescriptor(
         extension_id="coin-transfer",
         author=JWEBOT_AUTHOR,
-        description="金币转账",
+        description="摩拉转账",
         help_entries=("/转账 对方 金额；/transfer",),
         command_kinds=("transfer",),
     ),
@@ -182,6 +187,13 @@ LOCAL_EXTENSIONS = (
         help_entries=("/大富翁榜",),
         command_kinds=("coin_ranking",),
     ),
+    LocalExtensionDescriptor(
+        extension_id="coin-taffy",
+        author=JWEBOT_AUTHOR,
+        description="Taffy摩拉",
+        help_entries=("/Taffy摩拉（兼容 /Taffy金币）",),
+        command_kinds=("taffy_coins",),
+    ),
     LocalExtensionDescriptor(
         extension_id="coin-loan",
         author=JWEBOT_AUTHOR,

packages/jwebot-official-extensions/src/jwebot_official_extensions/local_commands.py

@@ -157,6 +157,7 @@ class LocalCommandEngine:
         if kind in {
             "wallet",
             "coin_ranking",
+            "taffy_coins",
             "transfer",
             "loan",
             "repay",
@@ -364,7 +365,7 @@ class LocalCommandEngine:
         return (f"人物：{self.person_display_label(person['display_name'])}；"
                 f"群内索引 {person['person_key'][:12]}；别名：{aliases or '未配置'}。\n"
                 f"基本属性：{facts or '暂无配置'}。\n"
-                "本人金币 /金币；个人记忆 /我的记忆。身份由本地昵称映射解析，无法读取微信原生账号ID。")
+                "本人摩拉 /摩拉；个人记忆 /我的记忆。身份由本地昵称映射解析，无法读取微信原生账号ID。")
 
     @staticmethod
     def service_state(service_name: str) -> str:

packages/jwebot-official-extensions/src/jwebot_official_extensions/player_games.py

@@ -16,6 +16,9 @@ MAX_LOTTERY_BET = 500
 DEFAULT_LOTTERY_BET = 10
 MIN_BLACKJACK_BET = 1
 MAX_BLACKJACK_BET = 500
+MIN_BLACKJACK_TABLE_BET = 1
+MAX_BLACKJACK_TABLE_BET = 10_000
+MAX_BLACKJACK_TABLE_PLAYERS = 8
 MIN_TRANSFER_AMOUNT = 1
 MAX_TRANSFER_AMOUNT = 100_000
 MIN_LOAN_AMOUNT = 1
@@ -23,6 +26,8 @@ MAX_LOAN_AMOUNT = 500
 GAME_TIMEOUT_SECONDS = 600
 GAME_HISTORY_LIMIT = 50
 GAME_HISTORY_PAGE_SIZE = 5
+COIN_EVENT_LIMIT = 50
+COIN_EVENT_DISPLAY_LIMIT = 10
 GAME_LABELS = {
     "prize_draw": "抽奖",
     "lottery": "彩票",
@@ -45,11 +50,52 @@ BLACKJACK_COMBINED_ACTION = re.compile(
     r"^(?:开始|start)(?P<bet>\d{1,3})?"
     r"(?:要牌|hit|停牌|stand)$"
 )
-TRANSFER_MARKED = re.compile(
+BLACKJACK_TABLE_BET = re.compile(
-    r"^(?:给|to)?\s*(?P<target>.+?)\s*"
+    r"^(?:放|下注|bet)(?P<bet>\d{1,5})$",
-    r"(?:金额|金币|coin|coins|:|：)\s*(?P<amount>\d{1,6})$",
     re.IGNORECASE,
 )
+TRANSFER_MARKED = re.compile(
+    r"^(?:给|to)?\s*(?P<target>.+?)\s*"
+    r"(?:金额|摩拉|金币|coin|coins|:|：)\s*(?P<amount>\d{1,6})$",
+    re.IGNORECASE,
+)
+
+
+def coin_event_reason(kind: str, outcome: str) -> str:
+    label = GAME_LABELS.get(kind, kind)
+    if kind == "blackjack":
+        if "赢" in outcome or "获胜" in outcome:
+            return f"{label} 获胜"
+        if "输" in outcome or "认输" in outcome or "爆" in outcome:
+            return f"{label} 失败"
+        if "平" in outcome:
+            return f"{label} 平局"
+    return f"{label} {outcome}".strip()
+
+
+def append_coin_event(
+    account: dict[str, Any],
+    delta: int,
+    reason: str,
+) -> None:
+    if delta == 0:
+        return
+    events = account.get("coin_events")
+    if not isinstance(events, list):
+        events = []
+    events = [
+        event for event in events
+        if isinstance(event, dict) and type(event.get("delta")) is int
+    ]
+    events.append(
+        {
+            "delta": delta,
+            "reason": str(reason)[:120],
+            "balance": int(account.get("balance", 0)),
+            "created_at": time.time(),
+        }
+    )
+    account["coin_events"] = events[-COIN_EVENT_LIMIT:]
 
 
 def normalize_player_account(account: dict[str, Any]) -> dict[str, Any]:
@@ -83,6 +129,33 @@ def normalize_player_account(account: dict[str, Any]) -> dict[str, Any]:
     account["game_history"] = [
         entry for entry in history if isinstance(entry, dict)
     ][-GAME_HISTORY_LIMIT:]
+    events = account.get("coin_events")
+    if not isinstance(events, list):
+        events = []
+        for entry in account["game_history"]:
+            try:
+                stake = int(entry.get("stake", 0))
+                payout = int(entry.get("payout", 0))
+            except (TypeError, ValueError):
+                continue
+            delta = payout - stake
+            if delta == 0:
+                continue
+            events.append(
+                {
+                    "delta": delta,
+                    "reason": coin_event_reason(
+                        str(entry.get("kind", "")),
+                        str(entry.get("outcome", "")),
+                    ),
+                    "balance": int(account["balance"]),
+                    "created_at": float(entry.get("completed_at", 0) or 0),
+                }
+            )
+    account["coin_events"] = [
+        event for event in events
+        if isinstance(event, dict) and type(event.get("delta")) is int
+    ][-COIN_EVENT_LIMIT:]
     return account
 
 
@@ -126,6 +199,40 @@ def append_game_record(
     account["game_history"] = history[-GAME_HISTORY_LIMIT:]
 
 
+def append_game_coin_event(
+    account: dict[str, Any],
+    *,
+    kind: str,
+    outcome: str,
+    stake: int,
+    payout: int,
+) -> None:
+    append_coin_event(
+        account,
+        payout - stake,
+        coin_event_reason(kind, outcome),
+    )
+
+
+def recent_coin_event_lines(
+    account: dict[str, Any],
+    *,
+    limit: int = COIN_EVENT_DISPLAY_LIMIT,
+) -> list[str]:
+    normalize_player_account(account)
+    events = sorted(
+        account["coin_events"],
+        key=lambda event: float(event.get("created_at", 0) or 0),
+        reverse=True,
+    )
+    lines: list[str] = []
+    for event in events[:limit]:
+        delta = int(event["delta"])
+        reason = str(event.get("reason", "账户变动")).strip() or "账户变动"
+        lines.append(f"{delta:+d}：{reason}")
+    return lines
+
+
 def card_label(card: int) -> str:
     return ("A", "2", "3", "4", "5", "6", "7", "8", "9", "10", "J", "Q", "K")[card % 13]
 
@@ -148,10 +255,11 @@ class PlayerGames:
     @staticmethod
     def rules() -> str:
         return (
-            "每人首次使用获 1000 虚拟金币；/金币 查看余额；/游戏记录 查看本人记录。\n"
+            "每人首次使用获 1000 虚拟摩拉；/摩拉 查看余额（兼容 /金币）；"
+            "/游戏记录 查看本人记录。\n"
             "转账：/转账 对方 金额，或 /transfer name amount；"
-            "对方需先用 /金币 初始化账户。\n"
+            "对方需先用 /摩拉 或 /金币 初始化账户。\n"
-            "贷款：/贷款 [1～500] 借入金币；/还款 [金额|全部] 归还，"
+            "贷款：/贷款 [1～500] 借入摩拉；/还款 [金额|全部] 归还，"
             "同一时间最多一笔未还贷款。\n"
             "抽奖：/抽奖 [2～100]，5%返5倍、20%返2倍，其余无返还。\n"
             "多人彩票：/彩票 开始；/彩票 报名 3 4 7 [1～500]；"
@@ -161,8 +269,11 @@ class PlayerGames:
             "看到牌后再单独发送 /21点 要牌|停牌|状态|结束"
             "（/blackjack hit|stand|status|end）；"
             "每人最多一局，10分钟无操作视为认输，结束不退下注。\n"
+            "多人21点：/21点 多人 启动 创建牌桌；/21点 加入；"
+            "主持人 /21点 多人 开始 后按提示 /21点 放200"
+            "（1～10000）；轮到自己 /21点 要牌|停牌。\n"
             "21点获胜返2倍、平局返下注、自然21点返2.5倍向下取整，庄家软17停牌。"
-            "以上均含本金；虚拟金币不能充值或兑换。"
+            "以上均含本金；虚拟摩拉不能充值或兑换。"
         )
 
     def handle(self, row: dict[str, Any], kind: str, arguments: str) -> str:
@@ -177,17 +288,28 @@ class PlayerGames:
             return self.group_lottery(row, arguments)
         if kind == "coin_ranking":
             return self.coin_ranking(row, arguments)
+        if kind == "taffy_coins":
+            return self.taffy_coins(row, arguments)
         if kind == "transfer":
             return self.transfer(row, arguments)
         if kind in {"loan", "repay"}:
             return self.loan_or_repay(row, kind, arguments)
         if kind == "wallet" and arguments:
-            return "用法：/金币（/wallet），查看本人的虚拟金币；/游戏帮助 查看规则"
+            return "用法：/摩拉（兼容 /金币、/wallet），查看本人的虚拟摩拉；/游戏帮助 查看规则"
         if kind == "prize_draw":
             if arguments and not re.fullmatch(r"\d{1,3}", arguments):
-                return "用法：/抽奖 [2～100金币]（/draw），默认10金币"
+                return "用法：/抽奖 [2～100摩拉]（/draw），默认10摩拉"
             bet = int(arguments or 10)
         elif kind == "blackjack":
+            table_command = self.parse_blackjack_table_command(arguments)
+            if table_command is not None:
+                table_action, table_bet = table_command
+                return self.group_blackjack(
+                    row,
+                    table_action,
+                    table_bet,
+                    arguments,
+                )
             compacted = re.sub(r"\s+", "", arguments.casefold()) or "状态"
             start = BLACKJACK_START.fullmatch(compacted)
             if start is not None:
@@ -209,16 +331,38 @@ class PlayerGames:
                     "用法：先 /21点 开始 [1～500]（金额是本局下注，默认50）；"
                     "开局后再单独发送 /21点 要牌|停牌|状态|结束；空格可省略"
                 )
+            group = str(row["group_fingerprint"])
+            if (
+                action in {"hit", "stand", "status"}
+                and self.ledger.has_active_group_blackjack(group)
+            ):
+                return self.group_blackjack(row, action, 0, arguments)
+            if (
+                action == "start"
+                and self.ledger.has_active_group_blackjack(group)
+            ):
+                return (
+                    "当前已有Taffy的21点多人牌桌；"
+                    "发送 /21点 状态 查看，或等待本桌结束"
+                )
+            if (
+                action == "end"
+                and self.ledger.has_active_group_blackjack(group)
+            ):
+                return (
+                    "当前是多人21点牌桌；轮到你时请 /21点 停牌，"
+                    "主持人可 /21点 多人 取消"
+                )
         else:
             bet = 0
         if kind == "prize_draw" and not MIN_DRAW_BET <= bet <= MAX_DRAW_BET:
-            return "抽奖下注必须为 2～100 的整数金币"
+            return "抽奖下注必须为 2～100 的整数摩拉"
         if (
             kind == "blackjack"
             and action == "start"
             and not MIN_BLACKJACK_BET <= bet <= MAX_BLACKJACK_BET
         ):
-            return "21点下注必须为 1～500 的整数金币"
+            return "21点下注必须为 1～500 的整数摩拉"
         person = self.ledger.resolve_person(
             str(row["group_fingerprint"]),
             str(row["sender"]),
@@ -248,6 +392,13 @@ class PlayerGames:
                     payout=0,
                     summary="10分钟无有效操作，按认输结束",
                 )
+                append_game_coin_event(
+                    account,
+                    kind="blackjack",
+                    outcome="超时认输",
+                    stake=int(game["bet"]),
+                    payout=0,
+                )
                 account["game"] = None
                 account["losses"] += 1
                 expired = "上一局已超时，按认输结束；"
@@ -255,11 +406,20 @@ class PlayerGames:
                 state = "进行中" if account.get("game") else "无进行中的牌局"
                 loan = account.get("loan")
                 loan_text = f"未还贷款 {loan['principal']}；" if loan else "无贷款；"
-                return (
+                total = account["wins"] + account["losses"] + account["ties"]
-                    f"{expired}金币 {account['balance']}；{loan_text}21点 {state}；"
+                win_rate = (account["wins"] / total * 100) if total else 0.0
-                    f"已结算胜 {account['wins']} / 负 {account['losses']} / 平 {account['ties']}。"
+                lines = [
-                    "首次赠送1000仅发一次；/游戏帮助 查看规则"
+                    f"{player_label}：{expired}当前摩拉 {account['balance']}；{loan_text}",
-                )
+                    (
+                        f"21点 {state}；已结算胜 {account['wins']} / "
+                        f"负 {account['losses']} / 平 {account['ties']}，"
+                        f"胜率 {win_rate:.2f}%。"
+                    ),
+                    "最近10次账户变动：",
+                ]
+                lines.extend(recent_coin_event_lines(account) or ["暂无"])
+                lines.append("/游戏帮助 查看规则")
+                return "\n".join(lines)
             if kind == "blackjack" and action in {"hit", "stand", "status", "end"}:
                 active = account.get("game")
                 index = int(active["index"]) if active else None
@@ -275,7 +435,7 @@ class PlayerGames:
                     else message
                 )
             if account["balance"] < bet:
-                message = expired + f"金币不足：余额 {account['balance']}，本次需要 {bet}"
+                message = expired + f"摩拉不足：余额 {account['balance']}，本次需要 {bet}"
                 return player_reply(message) if kind == "blackjack" else message
             account["balance"] -= bet
             if kind == "blackjack":
@@ -307,10 +467,17 @@ class PlayerGames:
                 payout=payout,
                 summary=f"抽奖结果 {result}",
             )
+            append_game_coin_event(
+                account,
+                kind="prize_draw",
+                outcome=result,
+                stake=bet,
+                payout=payout,
+            )
             return player_reply(
                 expired
                 + f"{result}；下注 {bet}，返还 {payout}，"
-                f"净变化 {payout - bet:+d}；金币 {account['balance']}",
+                f"净变化 {payout - bet:+d}；摩拉 {account['balance']}",
                 index,
             )
 
@@ -321,26 +488,471 @@ class PlayerGames:
     def coin_ranking(self, row: dict[str, Any], arguments: str) -> str:
         value = arguments.strip()
         if value and not re.fullmatch(r"\d{1,2}", value):
-            return "用法：/大富翁榜 或 /金币排行 [人数]"
+            return "用法：/大富翁榜、/摩拉排行 [人数] 或 /金币排行 [人数]"
         limit = int(value or 10)
         if not 1 <= limit <= 20:
-            return "金币排行人数必须为 1～20"
+            return "摩拉排行人数必须为 1～20"
         records = self.ledger.player_coin_ranking(
             str(row["group_fingerprint"]),
             limit=limit,
         )
         if not records:
-            return "当前群还没有金币账户；发送 /金币 初始化"
+            return "当前群还没有摩拉账户；发送 /摩拉 或 /金币 初始化"
         lines = ["大富翁榜："]
         for index, record in enumerate(records, start=1):
             loan = int(record.get("loan", 0))
             loan_text = f"（未还贷款 {loan}）" if loan else ""
             lines.append(
-                f"{index}. {record['display_name']} {record['balance']} 金币"
+                f"{index}. {record['display_name']} {record['balance']} 摩拉"
                 f"{loan_text}"
             )
         return "\n".join(lines)
 
+    def taffy_coins(self, row: dict[str, Any], arguments: str) -> str:
+        value = arguments.strip()
+        if value and not re.fullmatch(r"\d{1,2}", value):
+            return "用法：/Taffy摩拉 [人数]（兼容 /Taffy金币）"
+        limit = int(value or 8)
+        if not 1 <= limit <= 20:
+            return "Taffy摩拉人数必须为 1～20"
+        stats = self.ledger.player_taffy_coin_stats(
+            str(row["group_fingerprint"]),
+            limit=limit,
+        )
+        if int(stats["records"]) == 0:
+            return "Taffy摩拉：当前群暂无已结算游戏记录"
+        net = int(stats["taffy_net"])
+        result = f"Taffy累计赢 {net} 摩拉" if net >= 0 else f"Taffy累计倒贴 {-net} 摩拉"
+        lines = [
+            (
+                f"Taffy摩拉：已结算 {stats['records']} 局；"
+                f"玩家总下注 {stats['stake']}，总返还 {stats['payout']}；"
+                f"{result}"
+            )
+        ]
+        if stats["players"]:
+            lines.append("最大上贡者排行（含负数）：")
+        else:
+            lines.append("最大上贡者排行：暂无已结算玩家明细")
+        for index, record in enumerate(stats["players"], start=1):
+            player_net = int(record["taffy_net"])
+            if player_net > 0:
+                verdict = f"上贡 {player_net}"
+            elif player_net < 0:
+                verdict = f"从Taffy赢走 {-player_net}"
+            else:
+                verdict = "打平 0"
+            lines.append(
+                f"{index}. {record['display_name']}：{verdict}；"
+                f"下注 {record['stake']}，返还 {record['payout']}，"
+                f"{record['records']} 局"
+            )
+        return "\n".join(lines)
+
+    @staticmethod
+    def parse_blackjack_table_command(
+        arguments: str,
+    ) -> tuple[str, int] | None:
+        compacted = re.sub(r"\s+", "", arguments.casefold())
+        if not compacted:
+            return None
+        if compacted in {"多人", "牌桌", "table", "multiplayer"}:
+            return "help", 0
+        if compacted in {
+            "多人启动",
+            "多人开桌",
+            "多人创建",
+            "牌桌启动",
+            "牌桌开桌",
+            "tablestart",
+            "multiplayerstart",
+        }:
+            return "start", 0
+        if compacted in {"加入", "join"}:
+            return "join", 0
+        if compacted in {
+            "多人开始",
+            "多人发牌",
+            "开始多人",
+            "开桌",
+            "发牌",
+            "tabledeal",
+            "multiplayerdeal",
+        }:
+            return "begin", 0
+        if compacted in {
+            "多人取消",
+            "牌桌取消",
+            "取消牌桌",
+            "tablecancel",
+            "multiplayercancel",
+        }:
+            return "cancel", 0
+        bet = BLACKJACK_TABLE_BET.fullmatch(compacted)
+        if bet is not None:
+            return "bet", int(bet["bet"])
+        return None
+
+    @staticmethod
+    def blackjack_players_text(players: list[dict[str, Any]]) -> str:
+        names = [str(player["display_name"]) for player in players]
+        return "、".join(names) if names else "暂无"
+
+    @staticmethod
+    def blackjack_hand_text(hand: list[int]) -> str:
+        if not hand:
+            return "未发牌"
+        return (
+            f"{' '.join(card_label(card) for card in hand)}"
+            f"（{hand_value(hand)}点）"
+        )
+
+    @staticmethod
+    def blackjack_prompt(player: dict[str, Any]) -> str:
+        return (
+            f"请{player['display_name']} 放自己的筹码\n"
+            "输入/21点 放200（1～10000）"
+        )
+
+    def group_blackjack(
+        self,
+        row: dict[str, Any],
+        action: str,
+        amount: int,
+        arguments: str,
+    ) -> str:
+        group = str(row["group_fingerprint"])
+        sender = str(row["sender"])
+        event_id = str(row["event_id"])
+        command = f"blackjack-table:{arguments}"
+        if action == "help":
+            return (
+                "多人21点用法：/21点 多人 启动；/21点 加入；"
+                "主持人 /21点 多人 开始；按提示 /21点 放200"
+                "（1～10000）；轮到自己 /21点 要牌 或 /21点 停牌"
+            )
+        if action == "start":
+            return self.ledger.start_group_blackjack(
+                group,
+                sender,
+                event_id,
+                command,
+                max_players=MAX_BLACKJACK_TABLE_PLAYERS,
+                render=self.render_blackjack_table_start,
+            )
+        if action == "join":
+            return self.ledger.join_group_blackjack(
+                group,
+                sender,
+                event_id,
+                command,
+                max_players=MAX_BLACKJACK_TABLE_PLAYERS,
+                render=self.render_blackjack_table_join,
+            )
+        if action == "begin":
+            return self.ledger.begin_group_blackjack_betting(
+                group,
+                sender,
+                event_id,
+                command,
+                self.render_blackjack_table_begin,
+            )
+        if action == "cancel":
+            return self.ledger.cancel_group_blackjack(
+                group,
+                sender,
+                event_id,
+                command,
+                self.render_blackjack_table_cancel,
+            )
+        if action == "status":
+            return self.ledger.group_blackjack_status(
+                group,
+                sender,
+                event_id,
+                command,
+                self.render_blackjack_table_status,
+            )
+        if action == "bet":
+            if not MIN_BLACKJACK_TABLE_BET <= amount <= MAX_BLACKJACK_TABLE_BET:
+                return "21点多人牌桌筹码必须为 1～10000 的整数摩拉"
+            deck = list(range(52))
+            self.rng.shuffle(deck)
+            return self.ledger.place_group_blackjack_bet(
+                group,
+                sender,
+                event_id,
+                command,
+                amount,
+                deck,
+                self.render_blackjack_table_bet,
+            )
+        if action in {"hit", "stand"}:
+            return self.ledger.play_group_blackjack(
+                group,
+                sender,
+                event_id,
+                command,
+                action,
+                self.render_blackjack_table_action,
+            )
+        raise ValueError(f"unsupported blackjack table action: {action}")
+
+    def render_blackjack_table_start(self, data: dict[str, Any]) -> str:
+        status = data["status"]
+        if status == "private_game":
+            return (
+                f"{data['display_name']}：你已有单人21点，"
+                "请先 /21点 停牌 或 /21点 结束"
+            )
+        players = data.get("players", [])
+        prefix = (
+            "当前已有Taffy 的黑杰克Blackjack 21点牌桌"
+            if status == "already_open"
+            else "欢迎来到Taffy 的黑杰克Blackjack 21点牌桌"
+        )
+        return (
+            f"{prefix}\n"
+            "输入/21点 加入 加入当前游戏！"
+            f"目前玩家：{self.blackjack_players_text(players)}\n"
+            "主持人发送 /21点 多人 开始 后，"
+            "Taffy会按顺序喊每个人放筹码"
+        )
+
+    def render_blackjack_table_join(self, data: dict[str, Any]) -> str:
+        status = data["status"]
+        if status == "no_table":
+            return "当前没有21点多人牌桌；发送 /21点 多人 启动"
+        if status == "closed":
+            return "这桌已经停止加入；发送 /21点 状态 查看当前进度"
+        if status == "full":
+            return f"这桌已满，最多 {data['max_players']} 人"
+        if status == "private_game":
+            return (
+                f"{data['display_name']}：你已有单人21点，"
+                "请先 /21点 停牌 或 /21点 结束"
+            )
+        if status == "already_joined":
+            return (
+                f"{data['display_name']} 已经在这桌；"
+                f"目前玩家：{self.blackjack_players_text(data['players'])}"
+            )
+        return (
+            f"{data['display_name']} 已加入Taffy的21点牌桌；"
+            f"目前玩家：{self.blackjack_players_text(data['players'])}"
+        )
+
+    def render_blackjack_table_begin(self, data: dict[str, Any]) -> str:
+        status = data["status"]
+        if status == "no_table":
+            return "当前没有21点多人牌桌；发送 /21点 多人 启动"
+        if status == "not_host":
+            return f"只有本桌主持人 {data['host_display_name']} 可以结束加入"
+        if status == "not_joining":
+            return "这桌已经结束加入；发送 /21点 状态 查看当前进度"
+        if status == "not_enough":
+            return (
+                "至少需要 2 名玩家才能开始多人21点；"
+                f"目前玩家：{self.blackjack_players_text(data['players'])}"
+            )
+        return (
+            "加入结束，Taffy 按顺序收筹码。\n"
+            f"{self.blackjack_prompt(data['prompt_player'])}"
+        )
+
+    def render_blackjack_table_status(self, data: dict[str, Any]) -> str:
+        if data["status"] == "no_table":
+            return "当前没有21点多人牌桌；发送 /21点 多人 启动"
+        state = str(data["state"])
+        table_number = int(data["table_number"])
+        players = data["players"]
+        if state == "joining":
+            return (
+                f"21点牌桌第 {table_number} 桌正在加入；"
+                f"目前玩家：{self.blackjack_players_text(players)}\n"
+                "输入/21点 加入 加入当前游戏；"
+                "主持人 /21点 多人 开始"
+            )
+        if state == "betting":
+            prompt = next(
+                (
+                    player for player in players
+                    if str(player["person_key"]) == str(data["turn_person_key"])
+                ),
+                None,
+            )
+            lines = [f"21点牌桌第 {table_number} 桌正在下注："]
+            for player in players:
+                bet = int(player["bet"])
+                lines.append(
+                    f"{player['display_name']} #{player['player_game_index']}："
+                    f"{bet if bet else '未放'}"
+                )
+            if prompt is not None:
+                lines.append(self.blackjack_prompt(prompt))
+            return "\n".join(lines)
+        if state == "playing":
+            prompt = next(
+                (
+                    player for player in players
+                    if str(player["person_key"]) == str(data["turn_person_key"])
+                ),
+                None,
+            )
+            dealer = data.get("dealer", [])
+            dealer_text = (
+                f"{card_label(dealer[0])} + 暗牌"
+                if dealer
+                else "未发牌"
+            )
+            lines = [
+                f"21点牌桌第 {table_number} 桌正在出牌；Taffy {dealer_text}"
+            ]
+            for player in players:
+                lines.append(
+                    f"{player['display_name']} #{player['player_game_index']}："
+                    f"{self.blackjack_hand_text(player['hand'])}；"
+                    f"下注 {player['bet']}；{player['state']}"
+                )
+            if prompt is not None:
+                lines.append(
+                    f"请{prompt['display_name']} 操作："
+                    "/21点 要牌 或 /21点 停牌"
+                )
+            return "\n".join(lines)
+        return "当前没有进行中的21点多人牌桌"
+
+    def render_blackjack_table_bet(self, data: dict[str, Any]) -> str:
+        status = data["status"]
+        if status == "no_table":
+            return "当前没有21点多人牌桌；发送 /21点 多人 启动"
+        if status == "not_betting":
+            return "现在不是放筹码阶段；发送 /21点 状态 查看当前进度"
+        if status == "not_player":
+            return "你不在当前21点牌桌；下一桌可发送 /21点 加入"
+        if status == "wrong_turn":
+            return self.blackjack_prompt(data["prompt_player"])
+        if status == "insufficient":
+            return (
+                f"{data['display_name']}：摩拉不足，余额 {data['balance']}，"
+                f"本次需要 {data['bet']}"
+            )
+        if status == "bet_accepted":
+            return (
+                f"{data['display_name']} 放下 {data['bet']}；"
+                f"摩拉 {data['balance']}\n"
+                f"{self.blackjack_prompt(data['prompt_player'])}"
+            )
+        if status == "dealt":
+            return self.render_blackjack_table_dealt(data)
+        if status == "settled":
+            return self.render_blackjack_table_settlement(data)
+        raise ValueError(f"unsupported blackjack bet status: {status}")
+
+    def render_blackjack_table_dealt(self, data: dict[str, Any]) -> str:
+        dealer = data.get("dealer", [])
+        dealer_text = (
+            f"{card_label(dealer[0])} + 暗牌"
+            if dealer
+            else "未发牌"
+        )
+        lines = [
+            (
+                f"{data['display_name']} 放下 {data['bet']}；"
+                "全部筹码已确认。Taffy 发牌。"
+            ),
+            f"Taffy 明牌 {dealer_text}",
+        ]
+        for player in data["players"]:
+            lines.append(
+                f"{player['display_name']} #{player['player_game_index']}："
+                f"{self.blackjack_hand_text(player['hand'])}；"
+                f"下注 {player['bet']}"
+            )
+        prompt = data["prompt_player"]
+        lines.append(
+            f"请{prompt['display_name']} 操作：/21点 要牌 或 /21点 停牌"
+        )
+        return "\n".join(lines)
+
+    def render_blackjack_table_action(self, data: dict[str, Any]) -> str:
+        status = data["status"]
+        if status == "no_table":
+            return "当前没有21点多人牌桌；发送 /21点 多人 启动"
+        if status == "not_playing":
+            return "现在不是出牌阶段；发送 /21点 状态 查看当前进度"
+        if status == "not_player":
+            return "你不在当前21点牌桌；下一桌可发送 /21点 加入"
+        if status == "wrong_turn":
+            prompt = data["prompt_player"]
+            return (
+                f"现在轮到 {prompt['display_name']}；"
+                "/21点 要牌 或 /21点 停牌"
+            )
+        if status == "not_actionable":
+            return "你这一手已经结束；等待本桌结算"
+        if status == "settled":
+            return self.render_blackjack_table_settlement(data)
+        player = data["player"]
+        lines: list[str] = []
+        if status == "hit":
+            lines.append(
+                f"{player['display_name']} 要牌，拿到 "
+                f"{card_label(data['drawn_card'])}；"
+                f"当前 {self.blackjack_hand_text(player['hand'])}"
+            )
+        elif status == "busted":
+            lines.append(
+                f"{player['display_name']} 要牌，拿到 "
+                f"{card_label(data['drawn_card'])}；"
+                f"{self.blackjack_hand_text(player['hand'])} 爆牌"
+            )
+        elif status == "stood":
+            lines.append(
+                f"{player['display_name']} 停牌；"
+                f"{self.blackjack_hand_text(player['hand'])}"
+            )
+        else:
+            raise ValueError(f"unsupported blackjack action status: {status}")
+        prompt = data["prompt_player"]
+        lines.append(
+            f"请{prompt['display_name']} 操作：/21点 要牌 或 /21点 停牌"
+        )
+        return "\n".join(lines)
+
+    def render_blackjack_table_settlement(self, data: dict[str, Any]) -> str:
+        dealer = data.get("dealer", [])
+        dealer_text = self.blackjack_hand_text(dealer)
+        lines = [
+            (
+                "Taffy 的黑杰克Blackjack 21点牌桌"
+                f"第 {data['table_number']} 桌结算："
+            ),
+            f"Taffy 庄家 {dealer_text}",
+            "这一桌每个人的变化：",
+        ]
+        for player in data["players"]:
+            bet = int(player["bet"])
+            payout = int(player["payout"])
+            lines.append(
+                f"{player['display_name']} #{player['player_game_index']}："
+                f"{player['outcome']}；牌 {self.blackjack_hand_text(player['hand'])}；"
+                f"下注 {bet}，返还 {payout}，净变化 {payout - bet:+d}；"
+                f"摩拉 {player['balance']}"
+            )
+        return "\n".join(lines)
+
+    def render_blackjack_table_cancel(self, data: dict[str, Any]) -> str:
+        status = data["status"]
+        if status == "no_table":
+            return "当前没有21点多人牌桌"
+        if status == "not_host":
+            return f"只有本桌主持人 {data['host_display_name']} 可以取消"
+        return (
+            f"21点牌桌第 {data['table_number']} 桌已取消；"
+            f"{data['refunded']} 人的下注已退回"
+        )
+
     @staticmethod
     def parse_transfer(
         arguments: str,
@@ -376,11 +988,11 @@ class PlayerGames:
             return (
                 "用法：/转账 对方 金额，例如 /转账 Bob 100；"
                 "英文 /transfer Bob 100。金额范围 1～100000，"
-                "对方需先发送 /金币 初始化账户"
+                    "对方需先发送 /摩拉 或 /金币 初始化账户"
             )
         target, amount = parsed
         if not MIN_TRANSFER_AMOUNT <= amount <= MAX_TRANSFER_AMOUNT:
-            return "转账金额必须为 1～100000 的整数金币"
+            return "转账金额必须为 1～100000 的整数摩拉"
 
         def render(data: dict[str, Any]) -> str:
             source = data["source_display_name"]
@@ -389,18 +1001,18 @@ class PlayerGames:
                 return f"{source}：不能给自己转账"
             if data["status"] == "target_missing":
                 return (
-                    f"{source}：{target_name} 还没有金币账户，"
+                    f"{source}：{target_name} 还没有摩拉账户，"
-                    "请先让对方发送 /金币 初始化"
+                    "请先让对方发送 /摩拉 或 /金币 初始化"
                 )
             if data["status"] == "insufficient":
                 return (
-                    f"{source}：金币不足，余额 {data['source_balance']}，"
+                    f"{source}：摩拉不足，余额 {data['source_balance']}，"
                     f"本次需要 {data['amount']}"
                 )
             return (
-                f"{source}：已向 {target_name} 转账 {data['amount']} 金币；"
+                f"{source}：已向 {target_name} 转账 {data['amount']} 摩拉；"
-                f"本人金币 {data['source_balance']}；"
+                f"本人摩拉 {data['source_balance']}；"
-                f"{target_name} 金币 {data['target_balance']}"
+                f"{target_name} 摩拉 {data['target_balance']}"
             )
 
         return self.ledger.transfer_player_coins(
@@ -420,7 +1032,7 @@ class PlayerGames:
                 return "用法：/贷款 [1～500]；同一时间最多一笔未还贷款"
             amount = int(value)
             if not MIN_LOAN_AMOUNT <= amount <= MAX_LOAN_AMOUNT:
-                return "贷款金额必须为 1～500 的整数金币"
+                return "贷款金额必须为 1～500 的整数摩拉"
         else:
             amount = 0
             if value not in {"", "全部", "all"}:
@@ -428,7 +1040,7 @@ class PlayerGames:
                     return "用法：/还款 [金额|全部]"
                 amount = int(value)
                 if amount <= 0:
-                    return "还款金额必须为正整数金币"
+                    return "还款金额必须为正整数摩拉"
 
         group = str(row["group_fingerprint"])
         sender = str(row["sender"])
@@ -441,7 +1053,7 @@ class PlayerGames:
             if kind == "loan":
                 if loan:
                     return (
-                        f"{player_label}：已有未还贷款 {loan['principal']} 金币；"
+                        f"{player_label}：已有未还贷款 {loan['principal']} 摩拉；"
                         "请先 /还款 [金额|全部]"
                     )
                 account["balance"] += amount
@@ -449,9 +1061,10 @@ class PlayerGames:
                     "principal": amount,
                     "borrowed_at": time.time(),
                 }
+                append_coin_event(account, amount, "贷款到账")
                 return (
-                    f"{player_label}：贷款 {amount} 金币已到账；"
+                    f"{player_label}：贷款 {amount} 摩拉已到账；"
-                    f"当前金币 {account['balance']}；未还贷款 {amount}"
+                    f"当前摩拉 {account['balance']}；未还贷款 {amount}"
                 )
             if not loan:
                 return f"{player_label}：当前没有未还贷款"
@@ -461,13 +1074,14 @@ class PlayerGames:
             else:
                 pay = min(amount, debt)
             if pay <= 0:
-                return f"{player_label}：金币不足，无法还款；未还贷款 {debt}"
+                return f"{player_label}：摩拉不足，无法还款；未还贷款 {debt}"
             if pay > int(account["balance"]):
                 return (
-                    f"{player_label}：金币不足，余额 {account['balance']}，"
+                    f"{player_label}：摩拉不足，余额 {account['balance']}，"
                     f"本次还款需要 {pay}"
                 )
             account["balance"] -= pay
+            append_coin_event(account, -pay, "还款")
             debt -= pay
             if debt:
                 account["loan"] = {
@@ -475,13 +1089,13 @@ class PlayerGames:
                     "borrowed_at": float(loan.get("borrowed_at", time.time())),
                 }
                 return (
-                    f"{player_label}：已还款 {pay} 金币；"
+                    f"{player_label}：已还款 {pay} 摩拉；"
-                    f"剩余贷款 {debt}；当前金币 {account['balance']}"
+                    f"剩余贷款 {debt}；当前摩拉 {account['balance']}"
                 )
             account["loan"] = None
             return (
-                f"{player_label}：已还清贷款 {pay} 金币；"
+                f"{player_label}：已还清贷款 {pay} 摩拉；"
-                f"当前金币 {account['balance']}"
+                f"当前摩拉 {account['balance']}"
             )
 
         return self.ledger.apply_player_action(
@@ -547,7 +1161,7 @@ class PlayerGames:
         if len(set(digits)) != 3:
             raise ValueError("彩票的三个数字必须互不重复")
         if not MIN_LOTTERY_BET <= stake <= MAX_LOTTERY_BET:
-            raise ValueError("彩票下注必须为 1～500 的整数金币")
+            raise ValueError("彩票下注必须为 1～500 的整数摩拉")
         return tuple(sorted(digits)), stake
 
     def group_lottery(self, row: dict[str, Any], arguments: str) -> str:
@@ -618,7 +1232,7 @@ class PlayerGames:
                         f"命中 {entry['matches']} 个；下注 {entry['stake']}，"
                         f"返还 {entry['payout']}，净变化 "
                         f"{entry['payout'] - entry['stake']:+d}；"
-                        f"金币 {entry['balance']}"
+                        f"摩拉 {entry['balance']}"
                     )
                 return "\n".join(lines)
 
@@ -666,7 +1280,7 @@ class PlayerGames:
                 return "当前没有进行中的彩票；先发送 /彩票 开始"
             if data["status"] == "insufficient":
                 return (
-                    f"{data['display_name']}：金币不足，余额 {data['balance']}，"
+                    f"{data['display_name']}：摩拉不足，余额 {data['balance']}，"
                     f"报名需要 {stake}"
                 )
             action = "报名已更新" if data["status"] == "updated" else "报名成功"
@@ -674,7 +1288,7 @@ class PlayerGames:
                 f"{data['display_name']} #{data['player_game_index']}："
                 f"彩票第 {data['round_number']} 把{action}；号码 "
                 f"{' '.join(map(str, data['numbers']))}；下注 {data['stake']}；"
-                f"金币 {data['balance']}"
+                f"摩拉 {data['balance']}"
             )
 
         return self.ledger.register_group_lottery(
@@ -739,9 +1353,16 @@ class PlayerGames:
                 payout=0,
                 summary="主动结束牌局，不退下注",
             )
+            append_game_coin_event(
+                account,
+                kind="blackjack",
+                outcome="主动认输",
+                stake=bet,
+                payout=0,
+            )
             account["game"] = None
             account["losses"] += 1
-            return f"本人的21点已认输结束，不退下注；金币 {account['balance']}"
+            return f"本人的21点已认输结束，不退下注；摩拉 {account['balance']}"
         game["updated_at"] = time.time()
         if action == "hit":
             game["player"].append(game["deck"].pop())
@@ -781,7 +1402,14 @@ class PlayerGames:
                 f"玩家{player}点，庄家{dealer}点"
             ),
         )
+        append_game_coin_event(
+            account,
+            kind="blackjack",
+            outcome=outcome,
+            stake=int(bet),
+            payout=int(payout),
+        )
         account["game"] = None
         return (f"21点结算：{outcome}；你的牌 {' '.join(map(card_label, game['player']))}（{player}点），"
                 f"庄家 {' '.join(map(card_label, game['dealer']))}（{dealer}点）；"
-                f"下注 {bet}，返还 {payout}，净变化 {payout - bet:+d}；金币 {account['balance']}")
+                f"下注 {bet}，返还 {payout}，净变化 {payout - bet:+d}；摩拉 {account['balance']}")

packages/jwebot-sdk/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "jwebot-sdk"
-version = "1.0.0b1+1016"
+version = "1.0.4b1+1016"
 description = "Typed, capability-limited extension API for JWeBot plugins"
 requires-python = ">=3.9"
 dependencies = []

packages/jwebot-sdk/src/jwebot_sdk/__init__.py

@@ -38,4 +38,4 @@ __all__ = [
     "SingleGroupContext",
 ]
 
-__version__ = "1.0.0b1+1016"
+__version__ = "1.0.4b1+1016"

pyproject.toml

@@ -1,5 +1,5 @@
 [tool.jwebot]
-release = "beta 1.0.0.1016"
+release = "beta 1.0.4.1016"
 production_deployments_enabled_by_default = false
 
 [tool.jwebot.workspace]

tests/test_deployment_scripts.py

@@ -58,6 +58,7 @@ class DeploymentScriptTests(unittest.TestCase):
             "wechat_adaptive_gui_bridge.py",
             "jwebot_edge.py",
             "jwebot_edge_delivery.py",
+            "jwebot_release_update.py",
             "jwebot_config",
             "jwebot_sdk",
             "jwebot_kernel",
@@ -126,6 +127,9 @@ class DeploymentScriptTests(unittest.TestCase):
             'jwebot_log "runtime install ready"',
             'jwebot_log "edge active"',
             'jwebot_log "adaptive active"',
+            'jwebot_log "release settlement starting"',
+            '"${target_runtime}/jwebot_release_update.py"',
+            'jwebot_log "release settlement ready"',
             'jwebot_log "deployment ready version=',
             'cd "${target_home}"',
         ):
@@ -138,6 +142,14 @@ class DeploymentScriptTests(unittest.TestCase):
             source.count("xdotool"),
             source.count("/usr/bin/xdotool"),
         )
+        self.assertLess(
+            source.index("trap - EXIT HUP INT TERM"),
+            source.index('jwebot_log "release settlement starting"'),
+        )
+        self.assertLess(
+            source.index('jwebot_log "release settlement ready"'),
+            source.index('jwebot_log "deployment ready version='),
+        )
 
     def test_production_reports_bounded_progress(self):
         source = PRODUCTION_INSTALLER.read_text(encoding="utf-8")

tests/test_jwebot_edge.py

@@ -677,6 +677,23 @@ class LedgerAndApplicationTests(unittest.TestCase):
         self.application.deepseek.complete.assert_called_once()
         self.application.deliver.assert_called_once()
 
+    def test_semantic_duplicate_with_new_event_id_is_ignored(self):
+        first = self.application.accept(
+            {**event("event-1", "@JWeBot /金币"), "source": "rapidocr-local"}
+        )
+        duplicate = self.application.accept(
+            {**event("event-2", "@JWeBot /金币"), "source": "rapidocr-local"}
+        )
+
+        self.assertEqual(first["status"], "reply")
+        self.assertEqual(duplicate["status"], "ignored")
+        self.assertTrue(duplicate["duplicate"])
+        with self.application.ledger.lock:
+            count = self.application.ledger.connection.execute(
+                "SELECT COUNT(*) FROM events"
+            ).fetchone()[0]
+        self.assertEqual(count, 1)
+
     def test_failed_fast_ack_never_delivers_the_generated_reply(self):
         self.application.deepseek.complete = mock.Mock(return_value="最终回复喵")
         self.application.deliver = mock.Mock()
@@ -867,9 +884,9 @@ class LedgerAndApplicationTests(unittest.TestCase):
         self.assertEqual(version["route"], "local")
         self.assertIn("JWeBot 版本", version["text"])
         self.assertIn("kernel", version["text"])
-        self.assertIn("最新版本：beta 1.0.0.1016", version["text"])
+        self.assertIn("最新版本：beta 1.0.4.1016", version["text"])
         self.assertIn("更新时间：2026-10-16", version["text"])
-        self.assertIn("更新摘要：Promote JWeBot", version["text"])
+        self.assertIn("更新摘要：新增版本更新补偿流程", version["text"])
         self.assertEqual(debug["route"], "local")
         self.assertIn("调试埋点", debug["text"])
         self.assertEqual(balance["route"], "local")

tests/test_official_extension_inventory.py

@@ -92,6 +92,7 @@ class LocalExtensionInventoryTests(unittest.TestCase):
                 "coin": JWEBOT_AUTHOR,
                 "coin-transfer": JWEBOT_AUTHOR,
                 "coin-ranking": JWEBOT_AUTHOR,
+                "coin-taffy": JWEBOT_AUTHOR,
                 "coin-loan": JWEBOT_AUTHOR,
                 "random-number": JWEBOT_AUTHOR,
                 "draw-lot": JWEBOT_AUTHOR,

tests/test_release_061.py

@@ -39,7 +39,7 @@ class Release061Tests(unittest.TestCase):
         return json.loads(row[0])
 
     def test_aliases_are_corresponding_and_case_insensitive(self):
-        pairs = [("上下文", "Context"), ("金币", "wallet"), ("抽奖", "draw"),
+        pairs = [("上下文", "Context"), ("摩拉", "wallet"), ("抽奖", "draw"),
                  ("买彩票", "lottery"), ("21点", "blackjack"),
                  ("人物列表", "people"), ("人物", "person"), ("游戏帮助", "games")]
         for chinese, english in pairs:
@@ -82,18 +82,18 @@ class Release061Tests(unittest.TestCase):
         self.assertEqual(ledger.search_repeat_history("group-a", "好"), [])
 
     def test_initial_funds_aliases_groups_and_replay(self):
-        self.assertIn("金币 1000", self.command("/wallet", event_id="wallet-1"))
+        self.assertIn("摩拉 1000", self.command("/wallet", event_id="wallet-1"))
         with mock.patch.object(self.local.players.rng, "randrange", return_value=99):
             first = self.command("/draw 10", event_id="draw-1")
             replay = self.command("/draw 10", sender="a1", event_id="draw-1")
         self.assertEqual(first, replay)
         self.assertEqual(self.state("a1")["balance"], 990)
-        self.assertIn("金币 1000", self.command("/wallet", sender="Bob"))
+        self.assertIn("摩拉 1000", self.command("/wallet", sender="Bob"))
-        self.assertIn("金币 1000", self.command("/wallet", group="group-b"))
+        self.assertIn("摩拉 1000", self.command("/wallet", group="group-b"))
         self.engine.ledger.connection.close()
         self.engine = MODULE.EdgeApplication(config(self.root))
         self.local = self.engine.local
-        self.assertIn("金币 990", self.command("/wallet", event_id="restart"))
+        self.assertIn("摩拉 990", self.command("/wallet", event_id="restart"))
 
     def test_concurrent_replay_debits_once_and_rolls_back_failure(self):
         replies = []
@@ -123,7 +123,7 @@ class Release061Tests(unittest.TestCase):
         with mock.patch.object(self.local.players.rng, "randrange", return_value=99):
             for _ in range(10):
                 self.command("/draw 100")
-            self.assertIn("金币不足", self.command("/draw 2"))
+            self.assertIn("摩拉不足", self.command("/draw 2"))
         self.assertEqual(self.state()["balance"], 0)
 
     def test_blackjack_is_private_persistent_and_settles_exactly_once(self):

tests/test_release_062.py

@@ -119,8 +119,8 @@ class Release062BlackjackTests(unittest.TestCase):
         self.assertIn("分两步", reply)
         self.assertIn("100 是本局下注", reply)
         self.assertIn("再单独发送", reply)
-        wallet = self.command("/金币")
+        wallet = self.command("/摩拉")
-        self.assertIn("金币 1000", wallet)
+        self.assertIn("摩拉 1000", wallet)
         self.assertIn("无进行中的牌局", wallet)
 
 

tests/test_release_063.py

@@ -176,6 +176,61 @@ class Release063GameTests(unittest.TestCase):
         self.assertIn("下注 500，返还 500", history)
         self.assertIn("1～500", self.command("/21点开始501"))
 
+    def test_multiplayer_blackjack_table_guides_turns_and_settles(self):
+        self.assertIn("黑杰克Blackjack 21点牌桌", self.command("/21点 多人 启动"))
+        self.assertIn("目前玩家：alice", self.command("/21点 状态"))
+        self.assertIn("bob 已加入", self.command("/21点 加入", sender="Bob"))
+        self.assertIn("请alice 放自己的筹码", self.command("/21点 多人 开始"))
+
+        alice_bet = self.command("/21点 放200", event_id="alice-table-bet")
+        replay = self.command(
+            "/21点 放200",
+            sender="a1",
+            event_id="alice-table-bet",
+        )
+        self.assertEqual(replay, alice_bet)
+        self.assertIn("请bob 放自己的筹码", alice_bet)
+        self.assertEqual(self.state()["balance"], 800)
+
+        def table_deck(cards):
+            first = [12, 8, 9, 24, 19, 32, 25]
+            cards[:] = first + [card for card in cards if card not in first]
+
+        with mock.patch.object(
+            self.engine.local.players.rng,
+            "shuffle",
+            side_effect=table_deck,
+        ):
+            dealt = self.command("/21点 放100", sender="Bob")
+        self.assertIn("全部筹码已确认", dealt)
+        self.assertIn("alice #1：K Q（20点）；下注 200", dealt)
+        self.assertIn("bob #1：9 7（16点）；下注 100", dealt)
+        self.assertIn("请alice 操作", dealt)
+        self.assertEqual(self.state("Bob")["balance"], 900)
+
+        alice_stand = self.command("/21点 停牌")
+        self.assertIn("alice 停牌", alice_stand)
+        self.assertIn("请bob 操作", alice_stand)
+
+        settled = self.command("/21点 要牌", sender="Bob")
+        self.assertIn("这一桌每个人的变化", settled)
+        self.assertIn("alice #1：你赢了", settled)
+        self.assertIn("下注 200，返还 400，净变化 +200；摩拉 1200", settled)
+        self.assertIn("bob #1：你输了", settled)
+        self.assertIn("下注 100，返还 0，净变化 -100；摩拉 900", settled)
+        self.assertEqual(self.state()["balance"], 1200)
+        self.assertEqual(self.state("Bob")["balance"], 900)
+        self.assertIn("alice #1：21点 你赢了", self.command("/游戏记录"))
+        self.assertIn(
+            "bob #1：21点 你输了",
+            self.command("/游戏记录", sender="Bob"),
+        )
+        self.assertIn("+200：21点 获胜", self.command("/金币"))
+        self.assertIn(
+            "-100：21点 失败",
+            self.command("/金币", sender="Bob"),
+        )
+
     def test_prize_draw_has_indexed_history(self):
         with mock.patch.object(
             self.engine.local.players.rng,
@@ -192,12 +247,12 @@ class Release063GameTests(unittest.TestCase):
         self.assertIn("alice #1：抽奖 未中奖", history)
 
     def test_transfer_requires_initialized_target_and_replays_once(self):
-        self.assertIn("金币 1000", self.command("/金币"))
+        self.assertIn("摩拉 1000", self.command("/摩拉"))
         missing = self.command("/转账 Bob 100", event_id="missing-target")
-        self.assertIn("bob 还没有金币账户", missing)
+        self.assertIn("bob 还没有摩拉账户", missing)
         self.assertEqual(self.state()["balance"], 1000)
 
-        self.assertIn("金币 1000", self.command("/金币", sender="Bob"))
+        self.assertIn("摩拉 1000", self.command("/金币", sender="Bob"))
         first = self.command("/转账 Bob 250", event_id="transfer-1")
         replay = self.command(
             "/转账 Bob 250",
@@ -206,40 +261,56 @@ class Release063GameTests(unittest.TestCase):
         )
 
         self.assertEqual(replay, first)
-        self.assertIn("alice：已向 bob 转账 250 金币", first)
+        self.assertIn("alice：已向 bob 转账 250 摩拉", first)
         self.assertEqual(self.state()["balance"], 750)
         self.assertEqual(self.state("Bob")["balance"], 1250)
+        alice_wallet = self.command("/金币", event_id="wallet-after-transfer-a")
+        bob_wallet = self.command(
+            "/金币",
+            sender="Bob",
+            event_id="wallet-after-transfer-b",
+        )
+        self.assertIn("alice：当前摩拉 750；无贷款；", alice_wallet)
+        self.assertIn("最近10次账户变动：", alice_wallet)
+        self.assertIn("-250：转账给 bob", alice_wallet)
+        self.assertIn("+1000：首次初始化", alice_wallet)
+        self.assertIn("bob：当前摩拉 1250；无贷款；", bob_wallet)
+        self.assertIn("+250：转账 来自 alice", bob_wallet)
 
     def test_transfer_rejects_invalid_self_and_insufficient_cases(self):
-        self.assertIn("金币 1000", self.command("/金币"))
+        self.assertIn("摩拉 1000", self.command("/金币"))
         self.assertIn("转账金额必须为 1～100000", self.command("/转账 Bob 0"))
         self.assertIn("不能给自己转账", self.command("/转账 a1 100"))
-        self.assertIn("金币 1000", self.command("/金币", sender="Bob"))
+        self.assertIn("摩拉 1000", self.command("/金币", sender="Bob"))
-        self.assertIn("金币不足", self.command("/transfer 1001 Bob"))
+        self.assertIn("摩拉不足", self.command("/transfer 1001 Bob"))
         self.assertEqual(self.state()["balance"], 1000)
         self.assertEqual(self.state("Bob")["balance"], 1000)
 
     def test_loan_repay_and_replay_once(self):
-        self.assertIn("金币 1000", self.command("/金币"))
+        self.assertIn("摩拉 1000", self.command("/金币"))
         first = self.command("/贷款 300", event_id="loan-1")
         replay = self.command("/贷款 300", event_id="loan-1")
 
         self.assertEqual(replay, first)
-        self.assertIn("贷款 300 金币已到账", first)
+        self.assertIn("贷款 300 摩拉已到账", first)
         self.assertEqual(self.state()["balance"], 1300)
         self.assertEqual(self.state()["loan"]["principal"], 300)
         self.assertIn("未还贷款 300", self.command("/金币"))
         self.assertIn("已有未还贷款 300", self.command("/贷款 100"))
 
         partial = self.command("/还款 120")
-        self.assertIn("已还款 120 金币", partial)
+        self.assertIn("已还款 120 摩拉", partial)
         self.assertEqual(self.state()["balance"], 1180)
         self.assertEqual(self.state()["loan"]["principal"], 180)
 
         done = self.command("/还款 全部")
-        self.assertIn("已还清贷款 180 金币", done)
+        self.assertIn("已还清贷款 180 摩拉", done)
         self.assertEqual(self.state()["balance"], 1000)
         self.assertIsNone(self.state()["loan"])
+        wallet = self.command("/金币", event_id="wallet-after-repay")
+        self.assertIn("+300：贷款到账", wallet)
+        self.assertIn("-120：还款", wallet)
+        self.assertIn("-180：还款", wallet)
 
     def test_loan_rejects_invalid_amounts_and_empty_repay(self):
         self.assertIn("贷款金额必须为 1～500", self.command("/贷款 0"))
@@ -247,20 +318,43 @@ class Release063GameTests(unittest.TestCase):
         self.assertIn("当前没有未还贷款", self.command("/还款 全部"))
 
     def test_coin_ranking_lists_accounts_by_balance(self):
-        self.assertIn("金币 1000", self.command("/金币"))
+        self.assertIn("摩拉 1000", self.command("/金币"))
-        self.assertIn("金币 1000", self.command("/金币", sender="Bob"))
+        self.assertIn("摩拉 1000", self.command("/金币", sender="Bob"))
-        self.assertIn("金币 1000", self.command("/金币", sender="Carol"))
+        self.assertIn("摩拉 1000", self.command("/金币", sender="Carol"))
         self.command("/转账 Bob 250")
         self.command("/贷款 100", sender="Carol")
 
         ranking = self.command("/大富翁榜")
 
         self.assertIn("大富翁榜：", ranking)
-        self.assertIn("1. bob 1250 金币", ranking)
+        self.assertIn("1. bob 1250 摩拉", ranking)
-        self.assertIn("2. carol 1100 金币（未还贷款 100）", ranking)
+        self.assertIn("2. carol 1100 摩拉（未还贷款 100）", ranking)
-        self.assertIn("3. alice 750 金币", ranking)
+        self.assertIn("3. alice 750 摩拉", ranking)
         self.assertNotIn("Dave", ranking)
-        self.assertIn("金币排行人数必须为 1～20", self.command("/金币排行 0"))
+        self.assertIn("摩拉排行人数必须为 1～20", self.command("/摩拉排行 0"))
+
+    def test_taffy_coin_stats_sums_all_game_history(self):
+        with mock.patch.object(
+            self.engine.local.players.rng,
+            "randrange",
+            return_value=99,
+        ):
+            self.command("/抽奖10")
+        with mock.patch.object(
+            self.engine.local.players.rng,
+            "randrange",
+            return_value=0,
+        ):
+            self.command("/抽奖10", sender="Bob")
+
+        stats = self.command("/Taffy摩拉")
+
+        self.assertIn("玩家总下注 20，总返还 50", stats)
+        self.assertIn("Taffy累计倒贴 30 摩拉", stats)
+        self.assertIn("最大上贡者排行（含负数）：", stats)
+        self.assertIn("alice：上贡 10", stats)
+        self.assertIn("bob：从Taffy赢走 40", stats)
+        self.assertIn("Taffy摩拉人数必须为 1～20", self.command("/Taffy摩拉 0"))
 
     def test_open_round_and_history_survive_restart(self):
         root = Path(self.temp.name)

tests/test_release_update.py

@@ -0,0 +1,210 @@
+#!/usr/bin/env python3
+
+import importlib.util
+import json
+import sqlite3
+import sys
+import tempfile
+import unittest
+from datetime import datetime, timezone
+from pathlib import Path
+
+
+PROJECT_ROOT = Path(__file__).resolve().parents[1]
+MODULE_PATH = PROJECT_ROOT / "jwebot_release_update.py"
+SPEC = importlib.util.spec_from_file_location("jwebot_release_update", MODULE_PATH)
+MODULE = importlib.util.module_from_spec(SPEC)
+assert SPEC.loader is not None
+sys.modules[SPEC.name] = MODULE
+SPEC.loader.exec_module(MODULE)
+
+
+class ReleaseUpdateTests(unittest.TestCase):
+    release = "beta 1.0.4.1016"
+
+    def setUp(self):
+        self.temporary = tempfile.TemporaryDirectory()
+        self.root = Path(self.temporary.name)
+        self.database = self.root / "edge.sqlite3"
+        self.backups = self.root / "backups"
+        self.version = self.root / "VERSION"
+        self.changelog = self.root / "CHANGELOG.md"
+        self.delivery_helper = self.root / "jwebot_edge_delivery.py"
+        self.version.write_text(self.release + "\n", encoding="utf-8")
+        self.changelog.write_text(
+            (
+                "# Changelog\n\n"
+                f"## {self.release} - 2026-10-16\n\n"
+                "- 更新摩拉账户并修复重复回复问题。\n"
+                "- Internal implementation detail.\n"
+            ),
+            encoding="utf-8",
+        )
+        self.delivery_helper.write_text("# test helper\n", encoding="utf-8")
+        connection = sqlite3.connect(self.database)
+        connection.executescript(
+            """
+            CREATE TABLE metadata (
+                key TEXT PRIMARY KEY,
+                value TEXT NOT NULL,
+                updated_at TEXT NOT NULL
+            );
+            CREATE TABLE player_accounts (
+                group_fingerprint TEXT NOT NULL,
+                person_key TEXT NOT NULL,
+                state_json TEXT NOT NULL,
+                updated_at TEXT NOT NULL,
+                PRIMARY KEY (group_fingerprint, person_key)
+            );
+            """
+        )
+        for person_key, balance in (("person-a", 1000), ("person-b", 50)):
+            connection.execute(
+                "INSERT INTO player_accounts VALUES (?, ?, ?, ?)",
+                (
+                    "group-a",
+                    person_key,
+                    json.dumps({"balance": balance}),
+                    "2026-10-16T00:00:00Z",
+                ),
+            )
+        connection.commit()
+        connection.close()
+
+    def tearDown(self):
+        self.temporary.cleanup()
+
+    def account_states(self):
+        connection = sqlite3.connect(self.database)
+        try:
+            rows = connection.execute(
+                "SELECT state_json FROM player_accounts ORDER BY person_key"
+            ).fetchall()
+        finally:
+            connection.close()
+        return [json.loads(row[0]) for row in rows]
+
+    def announcement_marker(self):
+        key = MODULE.announcement_key(self.release, "group-a")
+        connection = sqlite3.connect(self.database)
+        try:
+            row = connection.execute(
+                "SELECT value FROM metadata WHERE key = ?",
+                (key,),
+            ).fetchone()
+        finally:
+            connection.close()
+        return json.loads(row[0]) if row else None
+
+    def test_changelog_first_bullet_is_the_announcement_summary(self):
+        self.changelog.write_text(
+            (
+                "# Changelog\n\n"
+                f"## {self.release} - 2026-10-16\n\n"
+                "- 第一行更新摘要，\n"
+                "  同一条摘要可以换行。\n"
+                "- 第二条不进入群公告。\n"
+            ),
+            encoding="utf-8",
+        )
+        summary = MODULE.read_changelog_summary(self.changelog, self.release)
+        self.assertEqual(summary, "第一行更新摘要， 同一条摘要可以换行。")
+        announcement = MODULE.build_announcement(self.release, summary)
+        self.assertIn(f"{self.release} 更新现已完成", announcement)
+        self.assertIn("更新补偿：摩拉 x200", announcement)
+        self.assertNotIn("第二条", announcement)
+
+    def test_release_update_grants_once_and_announces_once(self):
+        delivered = []
+
+        def deliverer(path, group, announcement, event_id):
+            delivered.append((path, group, announcement, event_id))
+
+        applied_at = datetime(2026, 10, 16, 8, 0, tzinfo=timezone.utc)
+        first = MODULE.run_release_update(
+            release_file=self.version,
+            changelog=self.changelog,
+            database_path=self.database,
+            backup_dir=self.backups,
+            delivery_helper=self.delivery_helper,
+            deliverer=deliverer,
+            applied_at=applied_at,
+        )
+        second = MODULE.run_release_update(
+            release_file=self.version,
+            changelog=self.changelog,
+            database_path=self.database,
+            backup_dir=self.backups,
+            delivery_helper=self.delivery_helper,
+            deliverer=deliverer,
+            applied_at=applied_at,
+        )
+
+        self.assertEqual(first["compensation"], "applied")
+        self.assertEqual(second["compensation"], "already_applied")
+        self.assertEqual(first["accounts"], 2)
+        self.assertEqual(len(delivered), 1)
+        self.assertEqual(delivered[0][1], "group-a")
+        self.assertIn("更新摩拉账户并修复重复回复问题", delivered[0][2])
+        self.assertEqual([state["balance"] for state in self.account_states()], [1200, 250])
+        for state in self.account_states():
+            matches = [
+                event
+                for event in state["coin_events"]
+                if event["delta"] == 200
+                and event["reason"] == f"{self.release} 版本更新补偿"
+            ]
+            self.assertEqual(len(matches), 1)
+            self.assertEqual(matches[0]["created_at"], applied_at.timestamp())
+        self.assertEqual(self.announcement_marker()["state"], "delivered")
+        self.assertEqual(len(list(self.backups.glob("*.sqlite3"))), 1)
+
+    def test_uncertain_announcement_never_duplicates_compensation(self):
+        def uncertain(*_args):
+            raise MODULE.AnnouncementUncertainError("uncertain")
+
+        with self.assertRaises(MODULE.AnnouncementUncertainError):
+            MODULE.run_release_update(
+                release_file=self.version,
+                changelog=self.changelog,
+                database_path=self.database,
+                backup_dir=self.backups,
+                delivery_helper=self.delivery_helper,
+                deliverer=uncertain,
+            )
+        self.assertEqual([state["balance"] for state in self.account_states()], [1200, 250])
+        self.assertEqual(self.announcement_marker()["state"], "uncertain")
+
+        called = []
+
+        def delivered(*args):
+            called.append(args)
+
+        with self.assertRaises(MODULE.AnnouncementUncertainError):
+            MODULE.run_release_update(
+                release_file=self.version,
+                changelog=self.changelog,
+                database_path=self.database,
+                backup_dir=self.backups,
+                delivery_helper=self.delivery_helper,
+                deliverer=delivered,
+            )
+        self.assertFalse(called)
+
+        result = MODULE.run_release_update(
+            release_file=self.version,
+            changelog=self.changelog,
+            database_path=self.database,
+            backup_dir=self.backups,
+            delivery_helper=self.delivery_helper,
+            deliverer=delivered,
+            retry_uncertain=True,
+        )
+        self.assertEqual(result["compensation"], "already_applied")
+        self.assertEqual(len(called), 1)
+        self.assertEqual([state["balance"] for state in self.account_states()], [1200, 250])
+        self.assertEqual(self.announcement_marker()["state"], "delivered")
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_wechat_adaptive_recorder.py

@@ -423,6 +423,31 @@ class LayoutScalingTests(unittest.TestCase):
         fast_ack_sender.assert_called_once()
         legacy.assert_not_called()
 
+    def test_ignored_duplicate_does_not_fall_back(self):
+        legacy = mock.Mock(return_value="legacy")
+        response = {
+            "status": "ignored",
+            "route": "local",
+            "execution_path": "local_command",
+            "duplicate": True,
+        }
+        with mock.patch.object(
+            MODULE.jwebot_edge,
+            "request_edge",
+            return_value=response,
+        ):
+            reply = MODULE.edge_router_reply(
+                {
+                    "event_id": "event-2",
+                    "group_fingerprint": "group-a",
+                    "text": "@JWeBot /金币",
+                },
+                legacy_router=legacy,
+            )
+
+        self.assertEqual(reply, "NO_REPLY")
+        legacy.assert_not_called()
+
     def test_hash_command_uses_legacy_router_when_edge_is_unavailable(self):
         legacy = mock.Mock(return_value="legacy")
         with mock.patch.object(

tests/test_wechat_local_ocr.py

@@ -4,6 +4,7 @@ import importlib.util
 import subprocess
 import sys
 import tempfile
+import types
 import unittest
 from contextlib import nullcontext
 from datetime import datetime, timezone
@@ -517,6 +518,64 @@ class GuiCoordinationTests(unittest.TestCase):
 
 
 class OcrFallbackTests(unittest.TestCase):
+    def test_ocr_engine_limits_onnxruntime_threads_by_default(self):
+        captured = {}
+
+        class FakeRapidOCR:
+            def __init__(self, **kwargs):
+                captured.update(kwargs)
+
+        fake_module = types.SimpleNamespace(RapidOCR=FakeRapidOCR)
+        fake_numpy = types.SimpleNamespace(array=lambda value: value)
+        with mock.patch.dict(
+            sys.modules,
+            {"rapidocr": fake_module, "numpy": fake_numpy},
+        ), mock.patch.dict(
+            MODULE.os.environ,
+            {
+                "WECHAT_OCR_ONNX_INTRA_OP_THREADS": "",
+                "WECHAT_OCR_ONNX_INTER_OP_THREADS": "",
+            },
+        ):
+            MODULE.OcrEngine()
+
+        self.assertEqual(
+            captured["params"],
+            {
+                "EngineConfig.onnxruntime.intra_op_num_threads": 1,
+                "EngineConfig.onnxruntime.inter_op_num_threads": 1,
+            },
+        )
+
+    def test_ocr_engine_allows_thread_limit_overrides(self):
+        captured = {}
+
+        class FakeRapidOCR:
+            def __init__(self, **kwargs):
+                captured.update(kwargs)
+
+        fake_module = types.SimpleNamespace(RapidOCR=FakeRapidOCR)
+        fake_numpy = types.SimpleNamespace(array=lambda value: value)
+        with mock.patch.dict(
+            sys.modules,
+            {"rapidocr": fake_module, "numpy": fake_numpy},
+        ), mock.patch.dict(
+            MODULE.os.environ,
+            {
+                "WECHAT_OCR_ONNX_INTRA_OP_THREADS": "2",
+                "WECHAT_OCR_ONNX_INTER_OP_THREADS": "3",
+            },
+        ):
+            MODULE.OcrEngine()
+
+        self.assertEqual(
+            captured["params"],
+            {
+                "EngineConfig.onnxruntime.intra_op_num_threads": 2,
+                "EngineConfig.onnxruntime.inter_op_num_threads": 3,
+            },
+        )
+
     def test_low_confidence_line_uses_detection_fallback(self):
         engine = MODULE.OcrEngine.__new__(MODULE.OcrEngine)
         bubble = MODULE.Bubble(50, 50, 132, 68, 5000)
@@ -630,6 +689,46 @@ class OcrCacheTests(unittest.TestCase):
         self.assertEqual(engine.last_cache_hits, 0)
         self.assertEqual(engine.last_cache_misses, 1)
 
+    def test_antialias_drift_reuses_visual_signature_cache(self):
+        engine = MODULE.OcrEngine.__new__(MODULE.OcrEngine)
+        bubble = MODULE.Bubble(50, 60, 132, 40, 5000)
+        first = MODULE.Image.new("RGB", (300, 180), (238, 238, 240))
+        second = MODULE.Image.new("RGB", (300, 180), (242, 242, 244))
+        first.putpixel((70, 75), (20, 20, 20))
+        second.putpixel((70, 75), (35, 35, 35))
+
+        with mock.patch.object(
+            engine,
+            "_recognize_bubble",
+            return_value=self.message(60),
+        ) as recognize:
+            engine.recognize_messages(first, [bubble])
+            messages, _ = engine.recognize_messages(second, [bubble])
+
+        self.assertEqual(recognize.call_count, 1)
+        self.assertEqual(len(messages), 1)
+        self.assertEqual(engine.last_cache_hits, 1)
+        self.assertEqual(engine.last_cache_misses, 0)
+
+    def test_unrecognized_bubble_reuses_negative_cache(self):
+        engine = MODULE.OcrEngine.__new__(MODULE.OcrEngine)
+        image = MODULE.Image.new("RGB", (300, 180), "white")
+        bubble = MODULE.Bubble(50, 60, 132, 40, 5000)
+
+        with mock.patch.object(
+            engine,
+            "_recognize_bubble",
+            return_value=None,
+        ) as recognize:
+            first_messages, _ = engine.recognize_messages(image, [bubble])
+            second_messages, _ = engine.recognize_messages(image, [bubble])
+
+        self.assertEqual(first_messages, [])
+        self.assertEqual(second_messages, [])
+        self.assertEqual(recognize.call_count, 1)
+        self.assertEqual(engine.last_cache_hits, 1)
+        self.assertEqual(engine.last_cache_misses, 0)
+
     def test_periodic_refresh_revalidates_cached_bubbles(self):
         engine = MODULE.OcrEngine.__new__(MODULE.OcrEngine)
         image = MODULE.Image.new("RGB", (300, 180), "white")

tools/check-local.sh

@@ -11,6 +11,7 @@ cd "$(dirname "$0")/.."
 python3 -m py_compile \
   jwebot_edge.py \
   jwebot_edge_delivery.py \
+  jwebot_release_update.py \
   wechat_local_ocr.py \
   wechat_adaptive_recorder.py \
   wechat_adaptive_gui_bridge.py \

wechat_adaptive_recorder.py

@@ -761,6 +761,18 @@ def edge_router_reply(
                 outcome="reply",
             )
             return reply
+    if response.get("status") == "ignored":
+        TELEMETRY.elapsed(
+            "recorder.edge.request",
+            started,
+            event_id=event_id,
+            group_id=group_id,
+            route=str(response.get("route", "")),
+            execution_path=str(response.get("execution_path", "")),
+            outcome="ignored",
+            duplicate=bool(response.get("duplicate")),
+        )
+        return "NO_REPLY"
     logging.warning(
         "edge router fallback event_hash=%s status=%s",
         hashlib.sha256(

wechat_local_ocr.py

@@ -67,6 +67,8 @@ DAMAGE_HEARTBEAT_SECONDS = 2.0
 CAPTURE_BUFFER_FRAMES = 12
 OCR_MESSAGE_CACHE_SIZE = 256
 OCR_CACHE_REFRESH_SCANS = 120
+DEFAULT_OCR_ONNX_INTRA_OP_THREADS = 1
+DEFAULT_OCR_ONNX_INTER_OP_THREADS = 1
 WINDOW_LAYOUT_SETTLE_SECONDS = 0.25
 FOLLOW_LATEST_SETTLE_SECONDS = 0.20
 FOLLOW_LATEST_MOUSE_X = 1432
@@ -224,6 +226,17 @@ def normalize_text(value: str) -> str:
     return re.sub(r"\s+", " ", value).strip()
 
 
+def env_int(name: str, default: int) -> int:
+    raw = os.environ.get(name, "").strip()
+    if not raw:
+        return default
+    try:
+        return int(raw)
+    except ValueError:
+        logging.warning("invalid integer environment value name=%s", name)
+        return default
+
+
 def bubble_sequence_signature(
     sender_crop: Image.Image,
     body_crop: Image.Image,
@@ -1099,10 +1112,21 @@ class OcrEngine:
         from rapidocr import RapidOCR
 
         self._numpy = np
-        self._engine = RapidOCR()
+        self._engine = RapidOCR(
+            params={
+                "EngineConfig.onnxruntime.intra_op_num_threads": env_int(
+                    "WECHAT_OCR_ONNX_INTRA_OP_THREADS",
+                    DEFAULT_OCR_ONNX_INTRA_OP_THREADS,
+                ),
+                "EngineConfig.onnxruntime.inter_op_num_threads": env_int(
+                    "WECHAT_OCR_ONNX_INTER_OP_THREADS",
+                    DEFAULT_OCR_ONNX_INTER_OP_THREADS,
+                ),
+            }
+        )
         self._message_cache: OrderedDict[
             str,
-            tuple[str, str, float, float, float],
+            tuple[str, str, float, float, float] | None,
         ] = OrderedDict()
         self._scan_count = 0
         self.last_cache_hits = 0
@@ -1178,7 +1202,7 @@ class OcrEngine:
 
     def _cache(self) -> OrderedDict[
         str,
-        tuple[str, str, float, float, float],
+        tuple[str, str, float, float, float] | None,
     ]:
         cache = getattr(self, "_message_cache", None)
         if cache is None:
@@ -1258,15 +1282,17 @@ class OcrEngine:
             bubble_crop = image.crop(
                 (bubble.x, bubble.y, bubble.right, bubble.bottom)
             )
-            key = self._cache_key(sender_crop, bubble_crop)
             sequence_signature = bubble_sequence_signature(
                 sender_crop,
                 bubble_crop,
             )
-            cached = cache.get(key)
+            key = sequence_signature
-            if cached is not None:
+            if key in cache:
+                cached = cache[key]
                 hits += 1
                 cache.move_to_end(key)
+                if cached is None:
+                    continue
                 messages.append(
                     IncomingMessage(
                         sender=cached[0],
@@ -1287,6 +1313,10 @@ class OcrEngine:
                 visual_signature=sequence_signature,
             )
             if message is None:
+                cache[key] = None
+                cache.move_to_end(key)
+                while len(cache) > OCR_MESSAGE_CACHE_SIZE:
+                    cache.popitem(last=False)
                 continue
             cache[key] = (
                 message.sender,