feat: add multiplayer lottery and game records

CHANGELOG.md

@@ -1,5 +1,58 @@
 # Changelog
 
+## alpha 0.6.3.1015
+
+- Replace instant personal lottery tickets with one persistent multiplayer
+  lottery round per group: start, register/update, status, draw, and cancel.
+- Treat each ticket as three distinct digits from 0-9. Sort digits before
+  storage so `347` and `743` represent the same selection.
+- Accept spaced and compact OCR forms. Ticket stakes are 1-500 coins with a
+  default of 10 when omitted.
+- Return 0%, 50%, 120%, or 300% of the stake for matching zero, one, two, or
+  three drawn digits. All settlement values remain integer coins.
+- Commit entry updates, stake differences, all-player payouts, cancellation
+  refunds, and event replay responses in serialized SQLite transactions.
+- Raise the blackjack stake range to 1-500 coins.
+- Add per-game personal sequence numbers and `/游戏记录` (`/gamehistory`) for
+  prize draws, lottery rounds, and blackjack hands.
+- Prefix game output with the configured player label and game number, such as
+  `Jacky #1：...`; aliases share the same counter and account.
+
+## alpha 0.6.2.1015
+
+- Bound direct DeepSeek SSE reads to 30 seconds of total wall time in addition
+  to the existing 18-second socket inactivity timeout.
+- Stop reading as soon as the SSE `[DONE]` marker arrives instead of waiting
+  for the upstream connection to close.
+- Prevent a trickling or unfinished upstream response from occupying both
+  generation workers for about 15 minutes and delaying later final replies.
+- Preserve fast ACK delivery, three-attempt retry behavior, two generation
+  workers, and strict final-reply ordering.
+- Treat `/21点 开始 100` as a 100-coin stake and accept OCR forms with missing
+  whitespace, such as `/21点开始100` and `/blackjackstart100`.
+- Keep blackjack play as explicit steps: start and deal first, then hit or
+  stand in a separate command. Combined forms return this workflow without
+  opening a game or deducting funds.
+- Prefix blackjack hands, status and settlement replies with the current
+  group-scoped player's configured display label, such as `jacky：...`.
+
+## alpha 0.6.1.1015
+
+- Add three reviewed plugins: local context summaries for 100/200/300 messages,
+  person and alias lookup, and single-player virtual coin games.
+- Provide corresponding Chinese and English commands, including `/context`,
+  `/people`, `/person`, `/wallet`, `/draw`, `/lottery`, and `/blackjack`.
+- Grant each group-scoped person 1000 virtual coins once; persist account state,
+  private blackjack decks, timeout/forfeit rules, and win/loss statistics.
+- Commit funds, game state and event replay responses in one SQLite transaction;
+  retries never debit twice, and aliases share the canonical person's account.
+- Reconcile the message image after a two-second idle wait so a missed final
+  XDamage notification does not strand an isolated message; drain buffered Xlib
+  events before waiting on the socket. Preserve visual deduplication and guards.
+- Keep the generated feature list within the sender's 900-character boundary.
+- Document local extractive-summary coverage, OCR identity limitations,
+  game payouts, persistence and rollback behavior.
+
 ## alpha 0.6.0.1015
 
 - Replace repository-local and hardcoded deployment values with independent,

README.md

@@ -8,7 +8,18 @@ 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：`alpha 0.6.0.1015`。
+当前版本：`alpha 0.6.3.1015`。
+
+0.6.1 新增本地上下文总结、人物与别名索引、单人虚拟金币游戏，并增加
+两秒画面复核以补抓遗漏的消息变化。命令和游戏管理规则见
+[0.6.1 插件说明](docs/local-plugins-0.6.1.md)。
+0.6.2 为 DeepSeek 流式回复增加 30 秒总时限，避免未结束的上游连接长期
+占满生成 worker；21 点开局金额明确作为本局下注，并兼容 OCR 丢失空格。
+21 点牌面和结算以当前玩家展示名开头。快速 ACK、重试和严格顺序投递
+规则保持不变。
+0.6.3 将彩票改为群内多人轮次，加入报名、状态、主持人开奖和取消；
+抽奖、彩票、21 点统一使用 `Jacky #N：...` 局号并支持 `/游戏记录`。
+完整玩法见[多人彩票与游戏记录](docs/group-lottery-0.6.3.md)。
 
 ### 运行流程
 
@@ -72,6 +83,13 @@ PATH="$PWD/.venv/bin:$PATH" ./tools/check-local.sh
 
 ```text
 /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）
+/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
+/游戏记录 [页码]（/gamehistory）
+/游戏帮助（/games）
 /丢骰子 [NdM]
 /猜数字 开始 [上限]
 /石头剪刀布 [开始|加入|出|状态|结算|结束]
@@ -240,7 +258,7 @@ visually confirmed GUI bridge.
 Current version:
 
 ```text
-alpha 0.6.0.1015
+alpha 0.6.3.1015
 ```
 
 ## Runtime Flow
@@ -337,6 +355,23 @@ the final separator shown below:
 /steam时长Dota2人=玩家名
 ```
 
+Version 0.6.1 adds local `/context 100|200|300` extractive summaries,
+`/people` and `/person` lookup, and `/wallet`, `/draw`, `/lottery`, `/blackjack`
+single-player virtual coin games. Chinese aliases share the same implementations.
+Accounts receive 1000 virtual coins once; funds, game state and event replay
+results commit together. OCR display aliases identify group-scoped people;
+native WeChat account IDs are unavailable. See
+[plugin rules and capture reconciliation](docs/local-plugins-0.6.1.md).
+Version 0.6.2 adds a 30-second total deadline to direct DeepSeek streams so an
+unfinished upstream response cannot occupy both generation workers for about
+15 minutes. Blackjack start amounts are the current round's stake; compact OCR
+forms are accepted, while hit/stand remain explicit follow-up commands.
+Blackjack hand and settlement replies start with the current player's configured
+display label. ACKs, retries, and ordered final delivery remain unchanged.
+Version 0.6.3 replaces instant lottery tickets with persistent multiplayer
+group rounds, adds indexed personal game history, and raises blackjack stakes
+to 1-500 coins. See [multiplayer lottery and game records](docs/group-lottery-0.6.3.md).
+
 Steam prefers private per-person SteamID mappings from the active profile and
 uses the corresponding group display IDs and aliases. Without a private roster,
 it discovers the configured account's public friend list at runtime. No real

VERSION

@@ -1 +1 @@
-alpha 0.6.0.1015
+alpha 0.6.3.1015

docs/command-routing.md

@@ -25,7 +25,8 @@ error. They never fall through to DeepSeek or OpenClaw.
 
 The local command path is:
 
-1. XDamage reports a relevant message-region change.
+1. XDamage reports a relevant message-region change, or a two-second idle
+   reconciliation capture detects a missed final drawing.
 2. OpenCV segments incoming bubbles without running OCR over the whole screen.
 3. The OCR engine reuses exact visual-fingerprint matches from a bounded LRU.
 4. Only new or changed sender/body crops use RapidOCR.
@@ -45,6 +46,10 @@ because each bubble position still produces its own `IncomingMessage`.
 Changed pixels force fresh OCR. The cache is bounded and periodically
 revalidated.
 
+For the three local plugins added in `alpha 0.6.1.1015`, including bilingual
+commands, person identity limitations and atomic single-player game settlement,
+see [local plugin rules](local-plugins-0.6.1.md).
+
 ## Bundled Plugin Registration
 
 Bundled command metadata has one source:

docs/group-lottery-0.6.3.md

@@ -0,0 +1,68 @@
+# alpha 0.6.3.1015 多人彩票与游戏记录
+
+## 一把彩票怎么玩
+
+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。
+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。
+5. 主持人也可在开奖前发送 `/彩票 取消`，所有报名者全额退款。
+
+命令关键词和数字之间的空格均可省略。`/买彩票` 和 `/lottery` 保留为
+对应别名。
+
+## 选号与报名
+
+- 每张票必须是三个互不重复的 0～9 数字。
+- 数字顺序不影响结果：`347`、`743` 和 `3 4 7` 都保存为 `3 4 7`。
+- 每人每把只有一张票。再次报名会更新号码和下注，不增加个人局号。
+- 更新到更高下注时只补扣差额；降低下注时退回差额。
+- 下注范围是 1～500，省略时为 10。
+- 没有进行中的彩票时不能报名；没有报名者时不能开奖。
+- 只有创建本轮的主持人可以开奖或取消，状态查询对群内所有人开放。
+
+支持的报名示例：
+
+```text
+/彩票 347
+/彩票 743 100
+/彩票 报名 3 4 7 100
+/彩票报名347金额100
+/彩票347100
+/lottery join 347 100
+```
+
+## 返还规则
+
+报名时先扣除下注。开奖号和玩家号码按集合比较，返还包含本金：
+
+| 命中数 | 返还 |
+|---|---:|
+| 0 个 | 0% |
+| 1 个 | 50% |
+| 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
+返还 120，净变化为 `+20`。
+
+## 个人局号与记录
+
+抽奖、彩票和 21 点分别维护本人的游戏局号。第一次参与某种游戏显示
+`Jacky #1：...`，下一把显示 `Jacky #2：...`。配置别名解析到同一人物时，
+共用同一账户、局号和历史。
+
+发送 `/游戏记录 [页码]`、`/个人记录 [页码]` 或
+`/gamehistory [page]` 查看最近记录，每页五条。记录包含游戏、个人局号、
+结果、下注、返还和净变化，最多保留最近 50 条。
+
+21 点下注范围同步调整为 1～500。`/21点 开始 100` 只表示本局下注 100；
+发牌后再单独发送 `/21点 要牌` 或 `/21点 停牌`。
+
+## 持久化与并发
+
+彩票轮次、报名和结算保存在 Edge SQLite。报名/改注、所有玩家的开奖返还、
+取消退款和事件重放响应都在 `BEGIN IMMEDIATE` 事务中提交。重复事件不会
+重复扣款、发奖或增加局号；进程重启后未开奖轮次和个人记录仍然保留。

docs/local-plugins-0.6.1.md

@@ -0,0 +1,107 @@
+# alpha 0.6.1.1015 本地插件与消息抓取
+
+本版本增加三个插件。全部由本机命令引擎执行，不调用 DeepSeek 或
+OpenClaw；中文、英文命令对应同一实现，英文不区分大小写，OCR 丢失
+命令与参数间的空格时仍可解析。正常使用时仍需 @ 当前配置的机器人。
+
+| 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 | `/抽奖 [金额]` | `/draw [金额]` |
+| single-player-games | `/彩票 开始/报名/当前状态/开奖/取消` | `/lottery start/join/status/draw/cancel` |
+| single-player-games | `/21点 开始 [金额]`（空格可省略） | `/blackjack start [金额]`（空格可省略） |
+| single-player-games | `/21点 要牌/停牌/状态/结束` | `/blackjack hit/stand/status/end` |
+| single-player-games | `/游戏记录 [页码]` | `/gamehistory [page]` |
+| single-player-games | `/游戏帮助` | `/games` |
+| person-directory | `/人物列表 [页码]`、`/别名列表 [页码]` | `/people [页码]` |
+| person-directory | `/人物 [名称或别名]`、`/个人属性 [名称或别名]` | `/person [名称或别名]` |
+
+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`。
+
+## 上下文总结
+
+默认选择最近 100 条，也可选择 200 或 300。查询限定当前群，排除当前
+请求事件，按观察时间和事件 ID 选出最近记录，再恢复时间顺序。
+不足指定数量时明确报告实际数量；没有记录时返回空历史提示。
+
+总结采用本地抽取式方法：统计参与人数、时间范围、重复话题，选出五条
+具有代表性的内容并按时间排列。它保留原话，不生成未经原文支持的决定
+或待办。覆盖范围是本机已经成功记录的文字消息，不包含图片、语音、
+未加载的微信历史或本机没观察到的消息。所有历史只保存在本机私有状态。
+独立的 `context_history` 保留短消息，避免复读排行的长度过滤影响最近
+100/200/300 条计数；启动时从现有复读历史补入旧记录，查询时从 OCR
+历史导入已识别发送人的文字。
+
+## 人物与账户索引
+
+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。
+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、
+截短的群内索引、别名和经过安全过滤的基本属性，不返回 SteamID 或密钥。
+`/人物列表` 每页五人。真实人物资料仍只存放于 `~/.config/jwebot/`，
+仓库没有生产人物或映射文件。
+
+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。用户不能通过游戏命令指定他人的账户，
+不能充值、转账、赠币或兑换真实资产。
+
+## 虚拟金币与游戏管理
+
+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、
+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。
+
+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。
+alpha 0.6.3 将旧的个人即时彩票替换为群内多人轮次。三位数字顺序无关，
+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，
+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，
+牌堆、双方手牌、下注和最后操作时间持久化。A 可作为 1 或 11；
+庄家软 17 停牌。开始后只公开一张庄家牌；要牌、停牌、认输和查询都
+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。
+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。
+
+十分钟无有效游戏操作的牌局在下一次账户操作时按认输结束；查询状态
+不延长有效期。主动结束同样按认输处理，不退下注。其他人可以各自
+独立玩，不能加入或结束你的牌局。旧的群游戏保持独立。
+
+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；
+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
+人物或命令则拒绝。事务失败全部回滚，避免扣款后丢失结果。新的独立
+指令会创建新的游戏操作，即使文字相同也不会被当成重试。
+数据库迁移仅新增表，旧版本可以忽略它们；运行文件回滚不会主动删除
+游戏数据。使用协调部署器回滚整库时，以部署前已验证备份为恢复基线。
+
+## 单条消息抓取修复
+
+独立观察证据表明，新气泡完成绘制后有时没有对应的有效 XDamage
+事件；先前截图仍是旧布局，后续消息的绘制才唤醒生产抓取。
+
+保留 XDamage 快速路径，同时将无有效变化时的复核等待窗口缩短到两秒。
+空闲超时后走相同截图、跟随最新消息、视觉哈希和缓冲去重流程。
+画面没变化时不进入重复 OCR；最终绘制漏事件时也能补抓。等待 X socket
+前先检查 Xlib 已缓冲事件，避免用户态已有事件却阻塞等待另一次通知。
+
+新增 `source=reconcile` 遥测用于识别补抓路径。两秒是等待窗口，
+不是端到端响应承诺；截图、OCR、生成和 GUI 投递仍有各自耗时。
+短周期复核增加空闲截图开销，上线后需观测实际负载。
+群指纹、窗口就绪、GUI 串行、视觉确认和投递去重规则保持有效。

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

@@ -0,0 +1,35 @@
+# single-player-games
+
+Author: `jwebot`
+
+虚拟金币账户、抽奖、群内多人彩票和单人21点
+
+Help:
+
+- `/金币 /wallet；/抽奖 [金额] /draw；/游戏记录 /gamehistory`
+- `/彩票 开始|报名|当前状态|开奖|取消；/lottery`
+- `/21点 开始 [1～500]（金额=本局下注，默认50，空格可省略）`
+- `/21点 要牌|停牌|状态|结束；/blackjack hit|stand|status|end`
+- `/游戏帮助；/games`
+
+Registered 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`
+
+The adapter has no direct Ledger, GUI, environment, shell, or network access.
+It delegates to the reviewed `single-player-games` Kernel service profile
+and returns a structured `ReplyPlan`.

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

@@ -0,0 +1,14 @@
+id = "single-player-games"
+author = "jwebot"
+description = "虚拟金币账户、抽奖、群内多人彩票和单人21点"
+help = ["/金币 /wallet；/抽奖 [金额] /draw；/游戏记录 /gamehistory", "/彩票 开始|报名|当前状态|开奖|取消；/lottery", "/21点 开始 [1～500]（金额=本局下注，默认50，空格可省略）", "/21点 要牌|停牌|状态|结束；/blackjack hit|stand|status|end", "/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"]
+storage = false
+network_profiles = []
+service_profiles = ["single-player-games"]

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

@@ -0,0 +1,37 @@
+"""Generated reviewed adapter for single-player-games."""
+
+from jwebot_sdk import Event, ExtensionResult, ReplyPlan, SingleGroupContext
+
+
+PROFILE = 'single-player-games'
+AUTHOR = 'jwebot'
+DESCRIPTION = '虚拟金币账户、抽奖、群内多人彩票和单人21点'
+HELP_ENTRIES = ('/金币 /wallet；/抽奖 [金额] /draw；/游戏记录 /gamehistory', '/彩票 开始|报名|当前状态|开奖|取消；/lottery', '/21点 开始 [1～500]（金额=本局下注，默认50，空格可省略）', '/21点 要牌|停牌|状态|结束；/blackjack hit|stand|status|end', '/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')
+
+
+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(),
+            },
+        )
+        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/single-player-games/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 single_player_games 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, '虚拟金币账户、抽奖、群内多人彩票和单人21点')
+        self.assertEqual(HELP_ENTRIES, ('/金币 /wallet；/抽奖 [金额] /draw；/游戏记录 /gamehistory', '/彩票 开始|报名|当前状态|开奖|取消；/lottery', '/21点 开始 [1～500]（金额=本局下注，默认50，空格可省略）', '/21点 要牌|停牌|状态|结束；/blackjack hit|stand|status|end', '/游戏帮助；/games'))
+        self.assertIn('/金币', registered)
+        self.assertEqual(len(registered), 15)
+
+
+if __name__ == "__main__":
+    unittest.main()

extensions/official/context-summary/README.md

@@ -0,0 +1,18 @@
+# context-summary
+
+Author: `official`
+
+提取当前群最近100、200或300条本地文字的上下文
+
+Help:
+
+- `/上下文 [100|200|300]；/context [100|200|300]`
+
+Registered commands:
+
+- `/上下文`
+- `/context`
+
+The adapter has no direct Ledger, GUI, environment, shell, or network access.
+It delegates to the reviewed `context-summary` Kernel service profile
+and returns a structured `ReplyPlan`.

extensions/official/context-summary/plugin.toml

@@ -0,0 +1,14 @@
+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"
+help = ["/上下文 [100|200|300]；/context [100|200|300]"]
+version = "1.0.0"
+api_version = "1"
+entrypoint = "context_summary:plugin"
+type = "command"
+
+[capabilities]
+commands = ["/上下文", "/context"]
+storage = false
+network_profiles = []
+service_profiles = ["context-summary"]

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

@@ -0,0 +1,37 @@
+"""Generated reviewed adapter for context-summary."""
+
+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'
+HELP_ENTRIES = ('/上下文 [100|200|300]；/context [100|200|300]',)
+COMMANDS = ('/上下文', '/context')
+
+
+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(),
+            },
+        )
+        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/official/context-summary/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 context_summary 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, '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(HELP_ENTRIES, ('/上下文 [100|200|300]；/context [100|200|300]',))
+        self.assertIn('/上下文', registered)
+        self.assertEqual(len(registered), 2)
+
+
+if __name__ == "__main__":
+    unittest.main()

extensions/official/person-directory/README.md

@@ -0,0 +1,22 @@
+# person-directory
+
+Author: `official`
+
+查询当前群已配置人物、基本属性和别名
+
+Help:
+
+- `/人物列表 [页码] /people；/人物 [名称或别名] /person`
+
+Registered commands:
+
+- `/人物列表`
+- `/别名列表`
+- `/people`
+- `/人物`
+- `/个人属性`
+- `/person`
+
+The adapter has no direct Ledger, GUI, environment, shell, or network access.
+It delegates to the reviewed `person-directory` Kernel service profile
+and returns a structured `ReplyPlan`.

extensions/official/person-directory/plugin.toml

@@ -0,0 +1,14 @@
+id = "person-directory"
+author = "official"
+description = "查询当前群已配置人物、基本属性和别名"
+help = ["/人物列表 [页码] /people；/人物 [名称或别名] /person"]
+version = "1.0.0"
+api_version = "1"
+entrypoint = "person_directory:plugin"
+type = "command"
+
+[capabilities]
+commands = ["/人物列表", "/别名列表", "/people", "/人物", "/个人属性", "/person"]
+storage = false
+network_profiles = []
+service_profiles = ["person-directory"]

extensions/official/person-directory/src/person_directory.py

@@ -0,0 +1,37 @@
+"""Generated reviewed adapter for person-directory."""
+
+from jwebot_sdk import Event, ExtensionResult, ReplyPlan, SingleGroupContext
+
+
+PROFILE = 'person-directory'
+AUTHOR = 'official'
+DESCRIPTION = '查询当前群已配置人物、基本属性和别名'
+HELP_ENTRIES = ('/人物列表 [页码] /people；/人物 [名称或别名] /person',)
+COMMANDS = ('/人物列表', '/别名列表', '/people', '/人物', '/个人属性', '/person')
+
+
+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(),
+            },
+        )
+        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/official/person-directory/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 person_directory 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, 'official')
+        self.assertEqual(DESCRIPTION, '查询当前群已配置人物、基本属性和别名')
+        self.assertEqual(HELP_ENTRIES, ('/人物列表 [页码] /people；/人物 [名称或别名] /person',))
+        self.assertIn('/人物列表', registered)
+        self.assertEqual(len(registered), 6)
+
+
+if __name__ == "__main__":
+    unittest.main()

packages/jwebot-config/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "jwebot-config"
-version = "0.6.0a0+1015"
+version = "0.6.3a0+1015"
 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__ = "0.6.0a0+1015"
+__version__ = "0.6.3a0+1015"

packages/jwebot-extension-host/pyproject.toml

@@ -4,12 +4,12 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "jwebot-extension-host"
-version = "0.6.0a0+1015"
+version = "0.6.3a0+1015"
 description = "Single-release, single-invocation process host for JWeBot extensions"
 requires-python = ">=3.9"
 dependencies = [
-  "jwebot-sdk==0.6.0a0+1015",
+  "jwebot-sdk==0.6.3a0+1015",
-  "jwebot-kernel==0.6.0a0+1015",
+  "jwebot-kernel==0.6.3a0+1015",
 ]
 
 [project.scripts]

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

@@ -16,4 +16,4 @@ __all__ = [
     "process_request",
 ]
 
-__version__ = "0.6.0a0+1015"
+__version__ = "0.6.3a0+1015"

packages/jwebot-kernel/pyproject.toml

@@ -4,13 +4,13 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "jwebot-kernel"
-version = "0.6.0a0+1015"
+version = "0.6.3a0+1015"
 description = "Stable routing and extension authority for JWeBot"
 requires-python = ">=3.9"
 dependencies = [
-  "jwebot-config==0.6.0a0+1015",
+  "jwebot-config==0.6.3a0+1015",
-  "jwebot-sdk==0.6.0a0+1015",
+  "jwebot-sdk==0.6.3a0+1015",
-  "jwebot-official-extensions==0.6.0a0+1015",
+  "jwebot-official-extensions==0.6.3a0+1015",
   "tomli>=2; python_version < '3.11'",
 ]
 

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

@@ -107,4 +107,4 @@ __all__ = [
     "verify_installed_release_signature",
 ]
 
-__version__ = "0.6.0a0+1015"
+__version__ = "0.6.3a0+1015"

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

@@ -68,6 +68,11 @@ from jwebot_official_extensions.steam import (
     SteamClient,
     SteamUnavailable,
 )
+from jwebot_official_extensions.player_games import (
+    append_game_record,
+    next_game_index,
+    normalize_player_account,
+)
 from jwebot_sdk import Event, PlanKind
 
 
@@ -92,6 +97,10 @@ MAX_REPLY_CHARS = 900
 MAX_CONTEXT_TURNS = 6
 MAX_OUTBOX_ATTEMPTS = 3
 DIRECT_CHAT_GENERATION_WORKERS = 2
+DIRECT_API_SOCKET_TIMEOUT_SECONDS = 18.0
+DIRECT_API_WALL_TIMEOUT_SECONDS = 30.0
+DIRECT_API_MAX_RESPONSE_BYTES = 2_000_000
+DIRECT_API_READ_CHUNK_BYTES = 65_536
 FAST_ACKNOWLEDGEMENT = "喵......"
 PROFILE_MIN_AUTOMATIC_OBSERVATIONS = 3
 PROFILE_MAX_FACTS = 12
@@ -626,6 +635,59 @@ def deepseek_response_content(raw: bytes) -> str:
     return str(content)
 
 
+def read_direct_api_response(
+    response: Any,
+    *,
+    wall_timeout_seconds: float = DIRECT_API_WALL_TIMEOUT_SECONDS,
+    max_bytes: int = DIRECT_API_MAX_RESPONSE_BYTES,
+) -> bytes:
+    """Read an SSE response with an absolute deadline, not an idle timeout."""
+
+    deadline = time.monotonic() + wall_timeout_seconds
+    chunks: list[bytes] = []
+    total = 0
+    stream_tail = b""
+    done_marker = re.compile(
+        rb"(?:^|\r?\n)data:\s*\[DONE\]\s*(?:\r?\n|$)"
+    )
+    socket_connection = getattr(
+        getattr(getattr(response, "fp", None), "raw", None),
+        "_sock",
+        None,
+    )
+    read_chunk = getattr(response, "read1", response.read)
+    while True:
+        remaining = deadline - time.monotonic()
+        if remaining <= 0:
+            raise DeepSeekUnavailable("direct API exceeded total timeout")
+        if socket_connection is not None:
+            socket_connection.settimeout(
+                max(
+                    0.001,
+                    min(DIRECT_API_SOCKET_TIMEOUT_SECONDS, remaining),
+                )
+            )
+        try:
+            chunk = read_chunk(
+                min(DIRECT_API_READ_CHUNK_BYTES, max_bytes - total + 1)
+            )
+        except (TimeoutError, OSError) as error:
+            raise DeepSeekUnavailable("direct API transport failure") from error
+        if not chunk:
+            break
+        if not isinstance(chunk, bytes):
+            raise DeepSeekUnavailable("direct API returned invalid bytes")
+        total += len(chunk)
+        if total > max_bytes:
+            raise DeepSeekUnavailable("direct API response exceeded size limit")
+        chunks.append(chunk)
+        stream_scan = stream_tail + chunk
+        if done_marker.search(stream_scan):
+            break
+        stream_tail = stream_scan[-64:]
+    return b"".join(chunks)
+
+
 def load_environment_file(path: Path = ENVIRONMENT_FILE) -> None:
     """Load the active generated profile, then the legacy compatibility file."""
 
@@ -941,6 +1003,73 @@ class Ledger:
                     observed_at TEXT NOT NULL,
                     created_at TEXT NOT NULL
                 );
+                CREATE TABLE IF NOT EXISTS context_history (
+                    event_id TEXT PRIMARY KEY,
+                    group_fingerprint TEXT NOT NULL,
+                    display_name TEXT NOT NULL,
+                    text TEXT NOT NULL,
+                    observed_at TEXT NOT NULL
+                );
+                CREATE INDEX IF NOT EXISTS context_history_group_time
+                    ON context_history(group_fingerprint, observed_at, event_id);
+                INSERT OR IGNORE INTO context_history
+                    SELECT event_id, group_fingerprint, display_name, text, observed_at
+                    FROM repeat_history;
+                CREATE TABLE IF NOT EXISTS 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)
+                );
+                CREATE TABLE IF NOT EXISTS player_actions (
+                    group_fingerprint TEXT NOT NULL,
+                    event_id TEXT NOT NULL,
+                    person_key TEXT NOT NULL,
+                    command TEXT NOT NULL,
+                    response_text TEXT NOT NULL,
+                    created_at TEXT NOT NULL,
+                    PRIMARY KEY (group_fingerprint, event_id)
+                );
+                CREATE TABLE IF NOT EXISTS group_lottery_rounds (
+                    group_fingerprint TEXT NOT NULL,
+                    round_number INTEGER NOT NULL,
+                    state TEXT NOT NULL,
+                    host_person_key TEXT NOT NULL,
+                    host_display_name TEXT NOT NULL,
+                    drawn_numbers_json TEXT NOT NULL DEFAULT '[]',
+                    created_at TEXT NOT NULL,
+                    updated_at TEXT NOT NULL,
+                    PRIMARY KEY (group_fingerprint, round_number)
+                );
+                CREATE TABLE IF NOT EXISTS group_lottery_entries (
+                    group_fingerprint TEXT NOT NULL,
+                    round_number INTEGER NOT NULL,
+                    person_key TEXT NOT NULL,
+                    display_name TEXT NOT NULL,
+                    player_game_index INTEGER NOT NULL,
+                    numbers_json TEXT NOT NULL,
+                    stake INTEGER NOT NULL,
+                    matches INTEGER NOT NULL DEFAULT 0,
+                    payout INTEGER NOT NULL DEFAULT 0,
+                    created_at TEXT NOT NULL,
+                    updated_at TEXT NOT NULL,
+                    PRIMARY KEY (
+                        group_fingerprint, round_number, person_key
+                    ),
+                    FOREIGN KEY (group_fingerprint, round_number)
+                        REFERENCES group_lottery_rounds (
+                            group_fingerprint, round_number
+                        )
+                        ON DELETE CASCADE
+                );
+                CREATE UNIQUE INDEX IF NOT EXISTS group_lottery_one_open
+                    ON group_lottery_rounds(group_fingerprint)
+                    WHERE state = 'open';
+                CREATE INDEX IF NOT EXISTS group_lottery_entries_round
+                    ON group_lottery_entries(
+                        group_fingerprint, round_number, created_at, person_key
+                    );
                 CREATE INDEX IF NOT EXISTS events_group_created
                     ON events(group_fingerprint, created_at);
                 CREATE INDEX IF NOT EXISTS outbox_due
@@ -2713,14 +2842,20 @@ class Ledger:
     ) -> bool:
         content = bounded_text(text, limit=1_500)
         repeat_key = repeat_text_key(content)
+        if not event_id or not group_fingerprint or not content:
+            return False
+        person = self.resolve_person(group_fingerprint, sender)
+        with self.lock:
+            self.connection.execute(
+                "INSERT OR IGNORE INTO context_history VALUES (?, ?, ?, ?, ?)",
+                (event_id, group_fingerprint, person_display_label(person["display_name"]),
+                 content, observed_at),
+            )
         if (
-            not event_id
+            not repeat_key
-            or not group_fingerprint
-            or not repeat_key
             or len(repeat_key) < 2
         ):
             return False
-        person = self.resolve_person(group_fingerprint, sender)
         with self.lock, self.connection:
             cursor = self.connection.execute(
                 """
@@ -2846,6 +2981,659 @@ class Ledger:
             ).fetchall()
         return [dict(row) for row in rows]
 
+    def recent_context_messages(
+        self, group: str, limit: int, *, exclude_event_id: str = "",
+    ) -> list[dict[str, Any]]:
+        if limit not in {100, 200, 300}:
+            raise LocalCommandError("上下文数量只支持 100、200、300")
+        with self.lock:
+            rows = self.connection.execute(
+                """
+                SELECT event_id, display_name, text, observed_at
+                FROM context_history
+                WHERE group_fingerprint = ? AND event_id != ?
+                ORDER BY observed_at DESC, event_id DESC LIMIT ?
+                """, (group, exclude_event_id, limit),
+            ).fetchall()
+        return [dict(row) for row in reversed(rows)]
+
+    def person_directory(self, group: str) -> list[dict[str, Any]]:
+        self.seed_person_aliases(group)
+        with self.lock:
+            rows = self.connection.execute(
+                """
+                SELECT canonical_name, alias_key, match_kind FROM person_aliases
+                WHERE group_fingerprint = ?
+                ORDER BY canonical_name, match_kind, alias_key
+                """, (group,),
+            ).fetchall()
+        people: dict[str, dict[str, Any]] = {}
+        for row in rows:
+            name = str(row["canonical_name"])
+            person = people.setdefault(name, {
+                "display_name": person_display_label(name),
+                "person_key": person_scope_key(group, name),
+                "aliases": [], "prefixes": [],
+            })
+            person["prefixes" if row["match_kind"] == "prefix" else "aliases"].append(
+                str(row["alias_key"]))
+        # Include profiles with a canonical name even when no alias was configured.
+        for template in PERSON_ALIAS_TEMPLATES:
+            name = str(template["canonical_name"])
+            people.setdefault(name, {
+                "display_name": person_display_label(name),
+                "person_key": person_scope_key(group, name),
+                "aliases": [], "prefixes": [],
+            })
+        return list(people.values())
+
+    def _player_identity(self, group: str, sender: str) -> dict[str, str]:
+        if not group or not sender or sender == "unknown":
+            raise LocalCommandError("无法确认本人身份，不能操作金币账户")
+        person = self.resolve_person(group, sender)
+        person["display_name"] = person_display_label(person["display_name"])
+        return person
+
+    def _player_action_replay(
+        self,
+        group: str,
+        event_id: str,
+        person_key: str,
+        command: str,
+    ) -> str | None:
+        previous = self.connection.execute(
+            """
+            SELECT person_key, command, response_text FROM player_actions
+            WHERE group_fingerprint = ? AND event_id = ?
+            """,
+            (group, event_id),
+        ).fetchone()
+        if previous is None:
+            return None
+        if previous["person_key"] != person_key or previous["command"] != command:
+            raise LocalCommandError("同一事件不能变更账户或游戏指令")
+        return str(previous["response_text"])
+
+    def _load_player_account(
+        self,
+        group: str,
+        person_key: str,
+    ) -> dict[str, Any]:
+        stored = self.connection.execute(
+            """
+            SELECT state_json FROM player_accounts
+            WHERE group_fingerprint = ? AND person_key = ?
+            """,
+            (group, person_key),
+        ).fetchone()
+        if stored is None:
+            return normalize_player_account({})
+        try:
+            account = json.loads(str(stored["state_json"]))
+        except json.JSONDecodeError as error:
+            raise LocalCommandError("金币账户状态无效") from error
+        if not isinstance(account, dict):
+            raise LocalCommandError("金币账户状态无效")
+        return normalize_player_account(account)
+
+    def _save_player_account(
+        self,
+        group: str,
+        person_key: str,
+        account: dict[str, Any],
+        timestamp: str,
+    ) -> None:
+        normalize_player_account(account)
+        if (
+            type(account["balance"]) is not int
+            or not 0 <= account["balance"] <= 1_000_000_000
+        ):
+            raise LocalCommandError("金币结算超出允许范围")
+        self.connection.execute(
+            """
+            INSERT INTO player_accounts VALUES (?, ?, ?, ?)
+            ON CONFLICT(group_fingerprint, person_key) DO UPDATE
+            SET state_json = excluded.state_json,
+                updated_at = excluded.updated_at
+            """,
+            (
+                group,
+                person_key,
+                json.dumps(account, ensure_ascii=False),
+                timestamp,
+            ),
+        )
+
+    def _store_player_action(
+        self,
+        group: str,
+        event_id: str,
+        person_key: str,
+        command: str,
+        response: str,
+        timestamp: str,
+    ) -> None:
+        self.connection.execute(
+            "INSERT INTO player_actions VALUES (?, ?, ?, ?, ?, ?)",
+            (group, event_id, person_key, command, response, timestamp),
+        )
+
+    def _open_group_lottery(self, group: str) -> sqlite3.Row | None:
+        return self.connection.execute(
+            """
+            SELECT * FROM group_lottery_rounds
+            WHERE group_fingerprint = ? AND state = 'open'
+            LIMIT 1
+            """,
+            (group,),
+        ).fetchone()
+
+    def _group_lottery_entries(
+        self,
+        group: str,
+        round_number: int,
+    ) -> list[dict[str, Any]]:
+        rows = self.connection.execute(
+            """
+            SELECT * FROM group_lottery_entries
+            WHERE group_fingerprint = ? AND round_number = ?
+            ORDER BY created_at, person_key
+            """,
+            (group, round_number),
+        ).fetchall()
+        entries: list[dict[str, Any]] = []
+        for row in rows:
+            entry = dict(row)
+            try:
+                numbers = json.loads(str(entry["numbers_json"]))
+            except json.JSONDecodeError:
+                numbers = []
+            entry["numbers"] = [
+                int(value)
+                for value in numbers
+                if type(value) is int and 0 <= value <= 9
+            ]
+            entries.append(entry)
+        return entries
+
+    def start_group_lottery(
+        self,
+        group: str,
+        sender: str,
+        event_id: str,
+        command: str,
+        render: Callable[[dict[str, Any]], str],
+    ) -> str:
+        person = self._player_identity(group, sender)
+        person_key = str(person["person_key"])
+        with self.lock:
+            self.connection.execute("BEGIN IMMEDIATE")
+            try:
+                replay = self._player_action_replay(
+                    group, event_id, person_key, command
+                )
+                if replay is not None:
+                    self.connection.execute("COMMIT")
+                    return replay
+                row = self._open_group_lottery(group)
+                status = "already_open"
+                if row is None:
+                    round_number = int(
+                        self.connection.execute(
+                            """
+                            SELECT COALESCE(MAX(round_number), 0) + 1
+                            FROM group_lottery_rounds
+                            WHERE group_fingerprint = ?
+                            """,
+                            (group,),
+                        ).fetchone()[0]
+                    )
+                    timestamp = utc_now()
+                    self.connection.execute(
+                        """
+                        INSERT INTO group_lottery_rounds (
+                            group_fingerprint, round_number, state,
+                            host_person_key, host_display_name,
+                            drawn_numbers_json, created_at, updated_at
+                        ) VALUES (?, ?, 'open', ?, ?, '[]', ?, ?)
+                        """,
+                        (
+                            group,
+                            round_number,
+                            person_key,
+                            person["display_name"],
+                            timestamp,
+                            timestamp,
+                        ),
+                    )
+                    row = self._open_group_lottery(group)
+                    status = "started"
+                if row is None:
+                    raise LocalCommandError("彩票创建失败")
+                data = dict(row)
+                data["status"] = status
+                response = render(data)
+                timestamp = utc_now()
+                self._store_player_action(
+                    group, event_id, person_key, command, response, timestamp
+                )
+                self.connection.execute("COMMIT")
+                return response
+            except BaseException:
+                self.connection.execute("ROLLBACK")
+                raise
+
+    def group_lottery_status(
+        self,
+        group: str,
+        sender: str,
+        event_id: str,
+        command: str,
+        render: Callable[[dict[str, Any]], str],
+    ) -> str:
+        person = self._player_identity(group, sender)
+        person_key = str(person["person_key"])
+        with self.lock:
+            self.connection.execute("BEGIN IMMEDIATE")
+            try:
+                replay = self._player_action_replay(
+                    group, event_id, person_key, command
+                )
+                if replay is not None:
+                    self.connection.execute("COMMIT")
+                    return replay
+                row = self._open_group_lottery(group)
+                if row is None:
+                    data: dict[str, Any] = {
+                        "status": "no_round",
+                        "entries": [],
+                    }
+                else:
+                    data = dict(row)
+                    data["status"] = "open"
+                    data["entries"] = self._group_lottery_entries(
+                        group, int(row["round_number"])
+                    )
+                response = render(data)
+                timestamp = utc_now()
+                self._store_player_action(
+                    group, event_id, person_key, command, response, timestamp
+                )
+                self.connection.execute("COMMIT")
+                return response
+            except BaseException:
+                self.connection.execute("ROLLBACK")
+                raise
+
+    def register_group_lottery(
+        self,
+        group: str,
+        sender: str,
+        event_id: str,
+        command: str,
+        numbers: tuple[int, int, int],
+        stake: int,
+        render: Callable[[dict[str, Any]], str],
+    ) -> str:
+        person = self._player_identity(group, sender)
+        person_key = str(person["person_key"])
+        with self.lock:
+            self.connection.execute("BEGIN IMMEDIATE")
+            try:
+                replay = self._player_action_replay(
+                    group, event_id, person_key, command
+                )
+                if replay is not None:
+                    self.connection.execute("COMMIT")
+                    return replay
+                row = self._open_group_lottery(group)
+                if row is None:
+                    data: dict[str, Any] = {
+                        "status": "no_round",
+                        "display_name": person["display_name"],
+                    }
+                else:
+                    round_number = int(row["round_number"])
+                    existing = self.connection.execute(
+                        """
+                        SELECT * FROM group_lottery_entries
+                        WHERE group_fingerprint = ? AND round_number = ?
+                          AND person_key = ?
+                        """,
+                        (group, round_number, person_key),
+                    ).fetchone()
+                    account = self._load_player_account(group, person_key)
+                    old_stake = int(existing["stake"]) if existing else 0
+                    delta = stake - old_stake
+                    if delta > int(account["balance"]):
+                        data = {
+                            "status": "insufficient",
+                            "display_name": person["display_name"],
+                            "balance": int(account["balance"]),
+                        }
+                    else:
+                        account["balance"] -= delta
+                        index = (
+                            int(existing["player_game_index"])
+                            if existing
+                            else next_game_index(account, "lottery")
+                        )
+                        timestamp = utc_now()
+                        self._save_player_account(
+                            group, person_key, account, timestamp
+                        )
+                        self.connection.execute(
+                            """
+                            INSERT INTO group_lottery_entries (
+                                group_fingerprint, round_number, person_key,
+                                display_name, player_game_index, numbers_json,
+                                stake, matches, payout, created_at, updated_at
+                            ) VALUES (?, ?, ?, ?, ?, ?, ?, 0, 0, ?, ?)
+                            ON CONFLICT (
+                                group_fingerprint, round_number, person_key
+                            ) DO UPDATE SET
+                                display_name = excluded.display_name,
+                                numbers_json = excluded.numbers_json,
+                                stake = excluded.stake,
+                                matches = 0,
+                                payout = 0,
+                                updated_at = excluded.updated_at
+                            """,
+                            (
+                                group,
+                                round_number,
+                                person_key,
+                                person["display_name"],
+                                index,
+                                canonical_json(list(numbers)),
+                                stake,
+                                timestamp,
+                                timestamp,
+                            ),
+                        )
+                        data = {
+                            "status": "updated" if existing else "registered",
+                            "display_name": person["display_name"],
+                            "player_game_index": index,
+                            "round_number": round_number,
+                            "numbers": list(numbers),
+                            "stake": stake,
+                            "balance": int(account["balance"]),
+                        }
+                response = render(data)
+                timestamp = utc_now()
+                self._store_player_action(
+                    group, event_id, person_key, command, response, timestamp
+                )
+                self.connection.execute("COMMIT")
+                return response
+            except BaseException:
+                self.connection.execute("ROLLBACK")
+                raise
+
+    def draw_group_lottery(
+        self,
+        group: str,
+        sender: str,
+        event_id: str,
+        command: str,
+        drawn_numbers: tuple[int, int, int],
+        render: Callable[[dict[str, Any]], str],
+    ) -> str:
+        person = self._player_identity(group, sender)
+        person_key = str(person["person_key"])
+        if (
+            len(drawn_numbers) != 3
+            or len(set(drawn_numbers)) != 3
+            or any(type(value) is not int or not 0 <= value <= 9 for value in drawn_numbers)
+        ):
+            raise LocalCommandError("彩票开奖号无效")
+        with self.lock:
+            self.connection.execute("BEGIN IMMEDIATE")
+            try:
+                replay = self._player_action_replay(
+                    group, event_id, person_key, command
+                )
+                if replay is not None:
+                    self.connection.execute("COMMIT")
+                    return replay
+                row = self._open_group_lottery(group)
+                if row is None:
+                    data: dict[str, Any] = {"status": "no_round"}
+                elif str(row["host_person_key"]) != person_key:
+                    data = {
+                        "status": "not_host",
+                        "host_display_name": str(row["host_display_name"]),
+                    }
+                else:
+                    round_number = int(row["round_number"])
+                    entries = self._group_lottery_entries(
+                        group, round_number
+                    )
+                    if not entries:
+                        data = {
+                            "status": "empty",
+                            "round_number": round_number,
+                        }
+                    else:
+                        timestamp = utc_now()
+                        settled: list[dict[str, Any]] = []
+                        drawn = set(drawn_numbers)
+                        rates = {0: 0, 1: 50, 2: 120, 3: 300}
+                        for entry in entries:
+                            key = str(entry["person_key"])
+                            account = self._load_player_account(group, key)
+                            matches = len(drawn.intersection(entry["numbers"]))
+                            stake = int(entry["stake"])
+                            payout = stake * rates[matches] // 100
+                            account["balance"] += payout
+                            account[
+                                "wins" if payout > stake else "losses"
+                            ] += 1
+                            append_game_record(
+                                account,
+                                kind="lottery",
+                                index=int(entry["player_game_index"]),
+                                outcome=f"命中{matches}个",
+                                stake=stake,
+                                payout=payout,
+                                summary=(
+                                    "选号 "
+                                    + " ".join(map(str, entry["numbers"]))
+                                    + "；开奖号 "
+                                    + " ".join(map(str, drawn_numbers))
+                                ),
+                            )
+                            self._save_player_account(
+                                group, key, account, timestamp
+                            )
+                            self.connection.execute(
+                                """
+                                UPDATE group_lottery_entries
+                                SET matches = ?, payout = ?, updated_at = ?
+                                WHERE group_fingerprint = ?
+                                  AND round_number = ? AND person_key = ?
+                                """,
+                                (
+                                    matches,
+                                    payout,
+                                    timestamp,
+                                    group,
+                                    round_number,
+                                    key,
+                                ),
+                            )
+                            result = dict(entry)
+                            result.update(
+                                {
+                                    "matches": matches,
+                                    "payout": payout,
+                                    "balance": int(account["balance"]),
+                                }
+                            )
+                            settled.append(result)
+                        self.connection.execute(
+                            """
+                            UPDATE group_lottery_rounds
+                            SET state = 'drawn', drawn_numbers_json = ?,
+                                updated_at = ?
+                            WHERE group_fingerprint = ? AND round_number = ?
+                            """,
+                            (
+                                canonical_json(list(drawn_numbers)),
+                                timestamp,
+                                group,
+                                round_number,
+                            ),
+                        )
+                        data = {
+                            "status": "drawn",
+                            "round_number": round_number,
+                            "drawn_numbers": list(drawn_numbers),
+                            "entries": settled,
+                        }
+                response = render(data)
+                timestamp = utc_now()
+                self._store_player_action(
+                    group, event_id, person_key, command, response, timestamp
+                )
+                self.connection.execute("COMMIT")
+                return response
+            except BaseException:
+                self.connection.execute("ROLLBACK")
+                raise
+
+    def cancel_group_lottery(
+        self,
+        group: str,
+        sender: str,
+        event_id: str,
+        command: str,
+        render: Callable[[dict[str, Any]], str],
+    ) -> str:
+        person = self._player_identity(group, sender)
+        person_key = str(person["person_key"])
+        with self.lock:
+            self.connection.execute("BEGIN IMMEDIATE")
+            try:
+                replay = self._player_action_replay(
+                    group, event_id, person_key, command
+                )
+                if replay is not None:
+                    self.connection.execute("COMMIT")
+                    return replay
+                row = self._open_group_lottery(group)
+                if row is None:
+                    data: dict[str, Any] = {"status": "no_round"}
+                elif str(row["host_person_key"]) != person_key:
+                    data = {
+                        "status": "not_host",
+                        "host_display_name": str(row["host_display_name"]),
+                    }
+                else:
+                    round_number = int(row["round_number"])
+                    entries = self._group_lottery_entries(
+                        group, round_number
+                    )
+                    timestamp = utc_now()
+                    for entry in entries:
+                        key = str(entry["person_key"])
+                        stake = int(entry["stake"])
+                        account = self._load_player_account(group, key)
+                        account["balance"] += stake
+                        append_game_record(
+                            account,
+                            kind="lottery",
+                            index=int(entry["player_game_index"]),
+                            outcome="本轮取消",
+                            stake=stake,
+                            payout=stake,
+                            summary="主持人取消，下注已全额退回",
+                        )
+                        self._save_player_account(
+                            group, key, account, timestamp
+                        )
+                    self.connection.execute(
+                        """
+                        UPDATE group_lottery_rounds
+                        SET state = 'cancelled', updated_at = ?
+                        WHERE group_fingerprint = ? AND round_number = ?
+                        """,
+                        (timestamp, group, round_number),
+                    )
+                    data = {
+                        "status": "cancelled",
+                        "round_number": round_number,
+                        "entries": entries,
+                    }
+                response = render(data)
+                timestamp = utc_now()
+                self._store_player_action(
+                    group, event_id, person_key, command, response, timestamp
+                )
+                self.connection.execute("COMMIT")
+                return response
+            except BaseException:
+                self.connection.execute("ROLLBACK")
+                raise
+
+    def player_game_history(
+        self,
+        group: str,
+        sender: str,
+    ) -> dict[str, Any]:
+        person = self._player_identity(group, sender)
+        with self.lock:
+            account = self._load_player_account(
+                group, str(person["person_key"])
+            )
+        history = sorted(
+            account["game_history"],
+            key=lambda entry: float(entry.get("completed_at", 0)),
+            reverse=True,
+        )
+        return {
+            "display_name": person["display_name"],
+            "history": history,
+            "game_counts": dict(account["game_counts"]),
+        }
+
+    def apply_player_action(
+        self, group: str, sender: str, event_id: str, command: str,
+        action: Callable[[dict[str, Any]], str],
+    ) -> str:
+        """Commit identity-scoped funds, game state and replay response together."""
+        person = self._player_identity(group, sender)
+        person_key = str(person["person_key"])
+        with self.lock:
+            self.connection.execute("BEGIN IMMEDIATE")
+            try:
+                replay = self._player_action_replay(
+                    group, event_id, person_key, command
+                )
+                if replay is not None:
+                    self.connection.execute("COMMIT")
+                    return replay
+                account = self._load_player_account(group, person_key)
+                response = action(account)
+                timestamp = utc_now()
+                self._save_player_account(
+                    group, person_key, account, timestamp
+                )
+                self._store_player_action(
+                    group,
+                    event_id,
+                    person_key,
+                    command,
+                    response,
+                    timestamp,
+                )
+                self.connection.execute("COMMIT")
+                return response
+            except BaseException:
+                self.connection.execute("ROLLBACK")
+                raise
+
     def status(self) -> dict[str, Any]:
         with self.lock:
             events = self.connection.execute("SELECT COUNT(*) FROM events").fetchone()[0]
@@ -3292,8 +4080,11 @@ class DeepSeekClient:
             method="POST",
         )
         try:
-            with urllib.request.urlopen(request, timeout=18) as response:
+            with urllib.request.urlopen(
-                raw = response.read(2_000_000)
+                request,
+                timeout=DIRECT_API_SOCKET_TIMEOUT_SECONDS,
+            ) as response:
+                raw = read_direct_api_response(response)
         except urllib.error.HTTPError as error:
             raise DeepSeekUnavailable(f"direct API HTTP {error.code}") from error
         except (urllib.error.URLError, TimeoutError, OSError) as error:

packages/jwebot-official-extensions/pyproject.toml

@@ -4,11 +4,11 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "jwebot-official-extensions"
-version = "0.6.0a0+1015"
+version = "0.6.3a0+1015"
 description = "Reviewed local command adapters and broker implementation for JWeBot"
 requires-python = ">=3.9"
 dependencies = [
-  "jwebot-sdk==0.6.0a0+1015",
+  "jwebot-sdk==0.6.3a0+1015",
 ]
 
 [tool.setuptools.packages.find]

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

@@ -62,4 +62,4 @@ __all__ = [
     "render_feature_list",
 ]
 
-__version__ = "0.6.0a0+1015"
+__version__ = "0.6.3a0+1015"

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

@@ -0,0 +1,37 @@
+"""Generated reviewed adapter for context-summary."""
+
+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'
+HELP_ENTRIES = ('/上下文 [100|200|300]；/context [100|200|300]',)
+COMMANDS = ('/上下文', '/context')
+
+
+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(),
+            },
+        )
+        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/person_directory.py

@@ -0,0 +1,37 @@
+"""Generated reviewed adapter for person-directory."""
+
+from jwebot_sdk import Event, ExtensionResult, ReplyPlan, SingleGroupContext
+
+
+PROFILE = 'person-directory'
+AUTHOR = 'official'
+DESCRIPTION = '查询当前群已配置人物、基本属性和别名'
+HELP_ENTRIES = ('/人物列表 [页码] /people；/人物 [名称或别名] /person',)
+COMMANDS = ('/人物列表', '/别名列表', '/people', '/人物', '/个人属性', '/person')
+
+
+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(),
+            },
+        )
+        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/single_player_games.py

@@ -0,0 +1,37 @@
+"""Generated reviewed adapter for single-player-games."""
+
+from jwebot_sdk import Event, ExtensionResult, ReplyPlan, SingleGroupContext
+
+
+PROFILE = 'single-player-games'
+AUTHOR = 'jwebot'
+DESCRIPTION = '虚拟金币账户、抽奖、群内多人彩票和单人21点'
+HELP_ENTRIES = ('/金币 /wallet；/抽奖 [金额] /draw；/游戏记录 /gamehistory', '/彩票 开始|报名|当前状态|开奖|取消；/lottery', '/21点 开始 [1～500]（金额=本局下注，默认50，空格可省略）', '/21点 要牌|停牌|状态|结束；/blackjack hit|stand|status|end', '/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')
+
+
+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(),
+            },
+        )
+        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/descriptors.py

@@ -38,6 +38,29 @@ LOCAL_COMMAND_ALIASES = {
     "debug": "telemetry_debug",
     "余额": "deepseek_balance",
     "steam": "steam_friends",
+    "上下文": "context_summary",
+    "context": "context_summary",
+    "金币": "wallet",
+    "wallet": "wallet",
+    "抽奖": "prize_draw",
+    "draw": "prize_draw",
+    "彩票": "lottery",
+    "买彩票": "lottery",
+    "lottery": "lottery",
+    "21点": "blackjack",
+    "二十一点": "blackjack",
+    "blackjack": "blackjack",
+    "游戏记录": "game_history",
+    "个人记录": "game_history",
+    "gamehistory": "game_history",
+    "游戏帮助": "game_rules",
+    "games": "game_rules",
+    "人物列表": "person_directory",
+    "别名列表": "person_directory",
+    "people": "person_directory",
+    "人物": "person_info",
+    "个人属性": "person_info",
+    "person": "person_info",
 }
 
 
@@ -91,6 +114,40 @@ DIRECT_KERNEL_COMMAND_KINDS = frozenset(
 )
 
 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",
+        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="虚拟金币账户、抽奖、群内多人彩票和单人21点",
+        help_entries=(
+            "/金币 /wallet；/抽奖 [金额] /draw；/游戏记录 /gamehistory",
+            "/彩票 开始|报名|当前状态|开奖|取消；/lottery",
+            "/21点 开始 [1～500]（金额=本局下注，默认50，空格可省略）",
+            "/21点 要牌|停牌|状态|结束；/blackjack hit|stand|status|end",
+            "/游戏帮助；/games",
+        ),
+        command_kinds=(
+            "wallet",
+            "prize_draw",
+            "lottery",
+            "blackjack",
+            "game_history",
+            "game_rules",
+        ),
+    ),
+    LocalExtensionDescriptor(
+        extension_id="person-directory",
+        author=OFFICIAL_AUTHOR,
+        description="查询当前群已配置人物、基本属性和别名",
+        help_entries=("/人物列表 [页码] /people；/人物 [名称或别名] /person",),
+        command_kinds=("person_directory", "person_info"),
+    ),
     LocalExtensionDescriptor(
         extension_id="person-memory",
         author=OFFICIAL_AUTHOR,
@@ -237,7 +294,6 @@ def render_feature_list() -> str:
         )
     for descriptor in LOCAL_EXTENSIONS:
         lines.append(
-            f"[{descriptor.author}/{descriptor.extension_id}] "
             f"{descriptor.description}："
             + "；".join(descriptor.help_entries)
         )

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

@@ -9,11 +9,13 @@ import json
 import re
 import subprocess
 import time
+from collections import Counter
 from datetime import datetime, timezone
 from typing import Any, Callable
 from zoneinfo import ZoneInfo
 
 from .descriptors import LOCAL_COMMAND_ALIASES, render_feature_list
+from .player_games import PlayerGames
 
 
 LOCAL_GAME_TIMEOUT_SECONDS = 10 * 60
@@ -104,6 +106,7 @@ class LocalCommandEngine:
         self.person_display_label = person_display_label
         self.profile_fact_is_safe = profile_fact_is_safe
         self.profile_expiry = profile_expiry
+        self.players = PlayerGames(ledger, person_display_label)
 
     def handle(self, row: dict[str, Any], event: dict[str, Any]) -> str:
         question = str(row["question"])
@@ -112,6 +115,19 @@ class LocalCommandEngine:
             raise LocalCommandError("本地命令不存在，发 /功能列表 查看可用本地功能")
         if kind == "feature_list":
             return self.reply_cleaner(self.feature_list())
+        if kind == "context_summary":
+            return self.reply_cleaner(self.context_summary(row, arguments))
+        if kind in {"person_directory", "person_info"}:
+            return self.reply_cleaner(self.person_index(row, kind, arguments))
+        if kind in {
+            "wallet",
+            "prize_draw",
+            "lottery",
+            "blackjack",
+            "game_history",
+            "game_rules",
+        }:
+            return self.reply_cleaner(self.players.handle(row, kind, arguments))
         if kind == "dice":
             return self.reply_cleaner(self.dice(row, arguments))
         if kind == "dice_challenge":
@@ -229,6 +245,82 @@ class LocalCommandEngine:
         )
         return f"本地记录命中“{query}”：{listing}"
 
+    def context_summary(self, row: dict[str, Any], arguments: str) -> str:
+        arguments = arguments.strip()
+        if arguments.strip() not in {"", "100", "200", "300"}:
+            return "用法：/上下文 100|200|300（/context），默认100条"
+        limit = int(arguments or 100)
+        self.ledger.bootstrap_repeat_history()
+        records = self.ledger.recent_context_messages(
+            str(row["group_fingerprint"]), limit,
+            exclude_event_id=str(row["event_id"]))
+        if not records:
+            return "当前群还没有可总结的本地文字记录"
+        # Local extractive summary: rank representative messages, retaining chronology.
+        stop = {"这个", "那个", "就是", "然后", "现在", "我们", "你们", "可以",
+                "不是", "一个", "什么", "怎么", "感觉", "还是", "已经", "没有",
+                "the", "and", "you", "that", "this", "jwebot"}
+        texts = [re.sub(r"^@\S+\s+", "", normalize_text(record["text"])) for record in records]
+        tokens = [set(re.findall(r"[a-zA-Z][a-zA-Z0-9_-]{2,}|[\u4e00-\u9fff]{2}", text.casefold())) - stop
+                  for text in texts]
+        frequency = Counter(token for message in tokens for token in message)
+        topics = [token for token, count in frequency.most_common(6) if count > 1]
+        scored = sorted(range(len(records)), key=lambda i: (
+            -sum(frequency[token] for token in tokens[i]),
+            -int(bool(re.search(r"决定|计划|问题|需要|建议|修复|明天|[?？]", texts[i]))),
+            -i))
+        selected: list[int] = []
+        seen: set[str] = set()
+        for index in scored:
+            if texts[index] in seen or texts[index].lstrip().startswith("/"):
+                continue
+            selected.append(index)
+            seen.add(texts[index])
+            if len(selected) == 5:
+                break
+        if not selected:
+            selected = list(range(max(0, len(records) - 3), len(records)))
+        people = len({record["display_name"] for record in records})
+        lines = [f"上下文总结（本地提取）：请求最近{limit}条，实际{len(records)}条，{people}人。",
+                 f"范围 {records[0]['observed_at'][:16]} ～ {records[-1]['observed_at'][:16]}。",
+                 "主要话题：" + ("、".join(topics) if topics else "话题分散，以下为代表性内容。")]
+        for index in sorted(selected):
+            record = records[index]
+            lines.append(f"• {record['display_name']}：{texts[index][:100]}")
+        lines.append("仅覆盖本机已记录的当前群文字；问题或计划按原话提取。")
+        return "\n".join(lines)
+
+    def person_index(self, row: dict[str, Any], kind: str, arguments: str) -> str:
+        group = str(row["group_fingerprint"])
+        directory = self.ledger.person_directory(group)
+        if kind == "person_directory":
+            if arguments and not re.fullmatch(r"[1-9]\d{0,2}", arguments):
+                return "用法：/人物列表 [页码]（/people），每页5人"
+            page = int(arguments or 1)
+            pages = max(1, (len(directory) + 4) // 5)
+            if page > pages:
+                return f"只有 {pages} 页；/人物列表 1"
+            lines = [f"当前群人物与别名（{page}/{pages}，共{len(directory)}人）"]
+            for person in directory[(page - 1) * 5:page * 5]:
+                aliases = "、".join(person["aliases"])[:80] or "未配置"
+                prefixes = "、".join(person["prefixes"])[:40]
+                lines.append(f"{person['display_name']}：{aliases}" + (f"；前缀 {prefixes}*" if prefixes else ""))
+            lines.append("/人物 [名称或别名]（/person）查看基本属性；/人物列表 页码 翻页")
+            return "\n".join(lines)
+        requested = normalize_text(arguments) or str(row["sender"])
+        person = self.ledger.resolve_person(group, requested)
+        entry = next((item for item in directory if item["person_key"] == person["person_key"]), None)
+        if arguments and entry is None:
+            return "当前群没有这个已配置人物或别名；/人物列表 查看索引"
+        facts = str(person.get("identity_fact", ""))
+        if facts and not self.profile_fact_is_safe(facts):
+            facts = ""
+        aliases = "、".join(entry["aliases"])[:160] if entry else "未配置"
+        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。")
+
     @staticmethod
     def service_state(service_name: str) -> str:
         try:

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

@@ -0,0 +1,578 @@
+"""Single-player virtual currency rules, invoked by the reviewed Kernel service."""
+
+from __future__ import annotations
+
+import random
+import re
+import time
+from typing import Any
+
+
+INITIAL_COINS = 1000
+MIN_DRAW_BET = 2
+MAX_DRAW_BET = 100
+MIN_LOTTERY_BET = 1
+MAX_LOTTERY_BET = 500
+DEFAULT_LOTTERY_BET = 10
+MIN_BLACKJACK_BET = 1
+MAX_BLACKJACK_BET = 500
+GAME_TIMEOUT_SECONDS = 600
+GAME_HISTORY_LIMIT = 50
+GAME_HISTORY_PAGE_SIZE = 5
+GAME_LABELS = {
+    "prize_draw": "抽奖",
+    "lottery": "彩票",
+    "blackjack": "21点",
+}
+BLACKJACK_ACTIONS = {
+    "要牌": "hit",
+    "hit": "hit",
+    "停牌": "stand",
+    "stand": "stand",
+    "状态": "status",
+    "status": "status",
+    "结束": "end",
+    "end": "end",
+}
+BLACKJACK_START = re.compile(
+    r"^(?:开始|start)(?P<bet>\d{1,3})?$"
+)
+BLACKJACK_COMBINED_ACTION = re.compile(
+    r"^(?:开始|start)(?P<bet>\d{1,3})?"
+    r"(?:要牌|hit|停牌|stand)$"
+)
+
+
+def normalize_player_account(account: dict[str, Any]) -> dict[str, Any]:
+    account.setdefault("balance", INITIAL_COINS)
+    account.setdefault("game", None)
+    account.setdefault("wins", 0)
+    account.setdefault("losses", 0)
+    account.setdefault("ties", 0)
+    counts = account.get("game_counts")
+    if not isinstance(counts, dict):
+        counts = {}
+        account["game_counts"] = counts
+    for kind in GAME_LABELS:
+        value = counts.get(kind, 0)
+        counts[kind] = value if type(value) is int and value >= 0 else 0
+    history = account.get("game_history")
+    if not isinstance(history, list):
+        history = []
+    account["game_history"] = [
+        entry for entry in history if isinstance(entry, dict)
+    ][-GAME_HISTORY_LIMIT:]
+    return account
+
+
+def next_game_index(account: dict[str, Any], kind: str) -> int:
+    normalize_player_account(account)
+    counts = account["game_counts"]
+    counts[kind] = int(counts.get(kind, 0)) + 1
+    return int(counts[kind])
+
+
+def append_game_record(
+    account: dict[str, Any],
+    *,
+    kind: str,
+    index: int,
+    outcome: str,
+    stake: int,
+    payout: int,
+    summary: str,
+) -> None:
+    normalize_player_account(account)
+    history = [
+        entry
+        for entry in account["game_history"]
+        if not (
+            entry.get("kind") == kind
+            and int(entry.get("index", 0)) == index
+        )
+    ]
+    history.append(
+        {
+            "kind": kind,
+            "index": index,
+            "outcome": outcome[:80],
+            "stake": stake,
+            "payout": payout,
+            "summary": summary[:240],
+            "completed_at": time.time(),
+        }
+    )
+    account["game_history"] = history[-GAME_HISTORY_LIMIT:]
+
+
+def card_label(card: int) -> str:
+    return ("A", "2", "3", "4", "5", "6", "7", "8", "9", "10", "J", "Q", "K")[card % 13]
+
+
+def hand_value(cards: list[int]) -> int:
+    total = sum(min(card % 13 + 1, 10) if card % 13 else 11 for card in cards)
+    aces = sum(card % 13 == 0 for card in cards)
+    while total > 21 and aces:
+        total -= 10
+        aces -= 1
+    return total
+
+
+class PlayerGames:
+    def __init__(self, ledger: Any, display_label: Any) -> None:
+        self.ledger = ledger
+        self.display_label = display_label
+        self.rng = random.SystemRandom()
+
+    @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"
+            "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]；"
+            "/彩票 当前状态；/彩票 开奖。三位号码顺序无关，省略金额默认10；"
+            "命中1/2/3个返下注的50%/120%/300%。\n"
+            "21点先用 /21点 开始 [1～500] 进场，金额是本局下注，默认50；空格可省略。\n"
+            "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"
+            "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、转账或兑换。"
+        )
+
+    def handle(self, row: dict[str, Any], kind: str, arguments: str) -> str:
+        arguments = arguments.strip()
+        if kind == "game_rules":
+            return self.rules()
+        if arguments.casefold() in {"帮助", "规则", "help", "rules"}:
+            return self.rules()
+        if kind == "game_history":
+            return self.game_history(row, arguments)
+        if kind == "lottery":
+            return self.group_lottery(row, 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"
+        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"
+            bet = int(arguments or 10)
+        elif kind == "blackjack":
+            compacted = re.sub(r"\s+", "", arguments.casefold()) or "状态"
+            start = BLACKJACK_START.fullmatch(compacted)
+            if start is not None:
+                action = "start"
+                bet = int(start["bet"] or 50)
+            else:
+                action = BLACKJACK_ACTIONS.get(compacted, "")
+                bet = 0
+            if not action:
+                combined = BLACKJACK_COMBINED_ACTION.fullmatch(compacted)
+                if combined is not None:
+                    amount = combined["bet"] or "50"
+                    return (
+                        "21点要分两步：先发送 /21点 开始 "
+                        f"{amount}（{amount} 是本局下注）；看到牌后再单独发送 "
+                        "/21点 要牌 或 /21点 停牌"
+                    )
+                return (
+                    "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"
+                )
+        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"
+        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"
+        person = self.ledger.resolve_person(
+            str(row["group_fingerprint"]),
+            str(row["sender"]),
+        )
+        player_label = self.display_label(person["display_name"])
+
+        def player_reply(text: str, index: int | None = None) -> str:
+            suffix = f" #{index}" if index is not None else ""
+            return f"{player_label}{suffix}：{text}"
+
+        def execute(account: dict[str, Any]) -> str:
+            normalize_player_account(account)
+            expired = ""
+            game = account.get("game")
+            if game and (
+                type(game.get("index")) is not int
+                or int(game["index"]) < 1
+            ):
+                game["index"] = next_game_index(account, "blackjack")
+            if game and time.time() - float(game["updated_at"]) >= GAME_TIMEOUT_SECONDS:
+                append_game_record(
+                    account,
+                    kind="blackjack",
+                    index=int(game["index"]),
+                    outcome="超时认输",
+                    stake=int(game["bet"]),
+                    payout=0,
+                    summary="10分钟无有效操作，按认输结束",
+                )
+                account["game"] = None
+                account["losses"] += 1
+                expired = "上一局已超时，按认输结束；"
+            if kind == "wallet":
+                state = "进行中" if account.get("game") else "无进行中的牌局"
+                return (f"{expired}金币 {account['balance']}；21点 {state}；"
+                        f"已结算胜 {account['wins']} / 负 {account['losses']} / 平 {account['ties']}。"
+                        "首次赠送1000仅发一次；/游戏帮助 查看规则")
+            if kind == "blackjack" and action in {"hit", "stand", "status", "end"}:
+                active = account.get("game")
+                index = int(active["index"]) if active else None
+                return player_reply(
+                    expired + self.play_blackjack(account, action),
+                    index,
+                )
+            if account.get("game"):
+                message = expired + "你已有一局21点，请先 /21点 停牌 或 /21点 结束"
+                return (
+                    player_reply(message, int(account["game"]["index"]))
+                    if kind == "blackjack"
+                    else message
+                )
+            if account["balance"] < bet:
+                message = expired + f"金币不足：余额 {account['balance']}，本次需要 {bet}"
+                return player_reply(message) if kind == "blackjack" else message
+            account["balance"] -= bet
+            if kind == "blackjack":
+                index = next_game_index(account, "blackjack")
+                deck = list(range(52))
+                self.rng.shuffle(deck)
+                game = {"bet": bet, "deck": deck[4:], "player": deck[:2],
+                        "dealer": deck[2:4], "updated_at": time.time(),
+                        "index": index}
+                account["game"] = game
+                if hand_value(game["player"]) == 21 or hand_value(game["dealer"]) == 21:
+                    return player_reply(
+                        expired + self.finish_blackjack(account, natural=True),
+                        index,
+                    )
+                return player_reply(expired + self.show_hand(game), index)
+            index = next_game_index(account, "prize_draw")
+            roll = self.rng.randrange(100)
+            payout = bet * (5 if roll < 5 else 2 if roll < 25 else 0)
+            result = "大奖" if roll < 5 else "小奖" if roll < 25 else "未中奖"
+            account["balance"] += payout
+            account["wins" if payout > bet else "losses"] += 1
+            append_game_record(
+                account,
+                kind="prize_draw",
+                index=index,
+                outcome=result,
+                stake=bet,
+                payout=payout,
+                summary=f"抽奖结果 {result}",
+            )
+            return player_reply(
+                expired
+                + f"{result}；下注 {bet}，返还 {payout}，"
+                f"净变化 {payout - bet:+d}；金币 {account['balance']}",
+                index,
+            )
+
+        return self.ledger.apply_player_action(
+            str(row["group_fingerprint"]), str(row["sender"]),
+            str(row["event_id"]), kind + ":" + arguments, execute)
+
+    @staticmethod
+    def parse_lottery_entry(
+        arguments: str,
+    ) -> tuple[tuple[int, int, int], int] | None:
+        value = arguments.strip().casefold()
+        for prefix in ("报名", "join"):
+            if value.startswith(prefix):
+                value = value[len(prefix):].strip()
+                break
+        if not value:
+            return None
+        marked = re.fullmatch(
+            r"([0-9])\s*([0-9])\s*([0-9])"
+            r"(?:金额|下注|bet)(\d{1,3})",
+            re.sub(r"[,，、]", "", value),
+        )
+        stake: int
+        if marked is not None:
+            digits = tuple(int(marked[index]) for index in range(1, 4))
+            stake = int(marked[4])
+        else:
+            tokens = re.sub(r"[,，、]", " ", value).split()
+            digits: tuple[int, int, int]
+            if len(tokens) == 1 and tokens[0].isdigit():
+                compact = tokens[0]
+                if not 3 <= len(compact) <= 6:
+                    return None
+                digits = tuple(int(item) for item in compact[:3])
+                stake = int(compact[3:] or DEFAULT_LOTTERY_BET)
+            elif (
+                len(tokens) == 2
+                and re.fullmatch(r"\d{3}", tokens[0])
+                and re.fullmatch(r"\d{1,3}", tokens[1])
+            ):
+                digits = tuple(int(item) for item in tokens[0])
+                stake = int(tokens[1])
+            elif (
+                len(tokens) in {3, 4}
+                and all(re.fullmatch(r"\d", item) for item in tokens[:3])
+                and (
+                    len(tokens) == 3
+                    or re.fullmatch(r"\d{1,3}", tokens[3])
+                )
+            ):
+                digits = tuple(int(item) for item in tokens[:3])
+                stake = (
+                    int(tokens[3])
+                    if len(tokens) == 4
+                    else DEFAULT_LOTTERY_BET
+                )
+            else:
+                return None
+        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")
+        return tuple(sorted(digits)), stake
+
+    def group_lottery(self, row: dict[str, Any], arguments: str) -> str:
+        group = str(row["group_fingerprint"])
+        sender = str(row["sender"])
+        event_id = str(row["event_id"])
+        command = "lottery:" + arguments
+        compacted = re.sub(r"\s+", "", arguments.casefold())
+
+        def status_text(data: dict[str, Any]) -> str:
+            if data["status"] == "no_round":
+                return "当前没有进行中的彩票；发送 /彩票 开始 创建新一把"
+            entries = data["entries"]
+            lines = [
+                f"彩票第 {data['round_number']} 把正在报名；"
+                f"主持人 {data['host_display_name']}；已报名 {len(entries)} 人"
+            ]
+            for entry in entries:
+                lines.append(
+                    f"{entry['display_name']} #{entry['player_game_index']}："
+                    f"{' '.join(map(str, entry['numbers']))}，下注 {entry['stake']}"
+                )
+            if not entries:
+                lines.append("暂无报名；/彩票 报名 3 4 7 [金额]")
+            return "\n".join(lines)
+
+        if compacted in {"开始", "start"}:
+            return self.ledger.start_group_lottery(
+                group,
+                sender,
+                event_id,
+                command,
+                lambda data: (
+                    f"彩票第 {data['round_number']} 把"
+                    + ("已经在报名中" if data["status"] == "already_open" else "已开始")
+                    + f"；主持人 {data['host_display_name']}。\n"
+                    "报名：/彩票 3 4 7 [金额]（金额1～500，默认10，空格可省略）；"
+                    "查询：/彩票 当前状态；主持人开奖：/彩票 开奖"
+                ),
+            )
+        if compacted in {"当前状态", "状态", "status"}:
+            return self.ledger.group_lottery_status(
+                group,
+                sender,
+                event_id,
+                command,
+                status_text,
+            )
+        if compacted in {"开奖", "draw"}:
+            drawn = tuple(sorted(self.rng.sample(range(10), 3)))
+
+            def draw_text(data: dict[str, Any]) -> str:
+                if data["status"] == "no_round":
+                    return "当前没有可以开奖的彩票"
+                if data["status"] == "not_host":
+                    return (
+                        f"只有本轮主持人 {data['host_display_name']} 可以开奖"
+                    )
+                if data["status"] == "empty":
+                    return "本轮还没有人报名，不能开奖"
+                lines = [
+                    f"彩票第 {data['round_number']} 把开奖号："
+                    f"{' '.join(map(str, data['drawn_numbers']))}"
+                ]
+                for entry in data["entries"]:
+                    lines.append(
+                        f"{entry['display_name']} #{entry['player_game_index']}："
+                        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']}"
+                    )
+                return "\n".join(lines)
+
+            return self.ledger.draw_group_lottery(
+                group,
+                sender,
+                event_id,
+                command,
+                drawn,
+                draw_text,
+            )
+        if compacted in {"取消", "cancel"}:
+            return self.ledger.cancel_group_lottery(
+                group,
+                sender,
+                event_id,
+                command,
+                lambda data: (
+                    "当前没有进行中的彩票"
+                    if data["status"] == "no_round"
+                    else (
+                        f"只有本轮主持人 {data['host_display_name']} 可以取消"
+                        if data["status"] == "not_host"
+                        else (
+                            f"彩票第 {data['round_number']} 把已取消；"
+                            f"{len(data['entries'])} 人的下注已全额退回"
+                        )
+                    )
+                ),
+            )
+        try:
+            parsed = self.parse_lottery_entry(arguments)
+        except ValueError as error:
+            return str(error)
+        if parsed is None:
+            return (
+                "用法：先 /彩票 开始；再 /彩票 报名 3 4 7 [金额]"
+                "（也可 /彩票347，金额1～500、默认10）；"
+                "/彩票 当前状态；主持人 /彩票 开奖"
+            )
+        numbers, stake = parsed
+
+        def register_text(data: dict[str, Any]) -> str:
+            if data["status"] == "no_round":
+                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"报名需要 {stake}"
+                )
+            action = "报名已更新" if data["status"] == "updated" else "报名成功"
+            return (
+                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']}"
+            )
+
+        return self.ledger.register_group_lottery(
+            group,
+            sender,
+            event_id,
+            command,
+            numbers,
+            stake,
+            register_text,
+        )
+
+    def game_history(self, row: dict[str, Any], arguments: str) -> str:
+        if arguments and not re.fullmatch(r"\d{1,3}", arguments):
+            return "用法：/游戏记录 [页码]（/gamehistory），每页5条"
+        page = int(arguments or 1)
+        if page < 1:
+            return "游戏记录页码必须从 1 开始"
+        data = self.ledger.player_game_history(
+            str(row["group_fingerprint"]),
+            str(row["sender"]),
+        )
+        records = data["history"]
+        pages = max(1, (len(records) + GAME_HISTORY_PAGE_SIZE - 1) // GAME_HISTORY_PAGE_SIZE)
+        if page > pages:
+            return f"{data['display_name']} 只有 {pages} 页游戏记录"
+        start = (page - 1) * GAME_HISTORY_PAGE_SIZE
+        lines = [f"{data['display_name']} 的游戏记录 {page}/{pages}："]
+        for record in records[start:start + GAME_HISTORY_PAGE_SIZE]:
+            label = GAME_LABELS.get(str(record["kind"]), str(record["kind"]))
+            lines.append(
+                f"{data['display_name']} #{record['index']}：{label} {record['outcome']}；"
+                f"下注 {record['stake']}，返还 {record['payout']}，"
+                f"净变化 {record['payout'] - record['stake']:+d}"
+            )
+        if not records:
+            lines.append("暂无已结算记录")
+        return "\n".join(lines)
+
+    @staticmethod
+    def show_hand(game: dict[str, Any]) -> str:
+        player = " ".join(card_label(card) for card in game["player"])
+        return (f"21点：你的牌 {player}（{hand_value(game['player'])}点）；"
+                f"庄家 {card_label(game['dealer'][0])} + 暗牌；下注 {game['bet']}。\n"
+                "/21点 要牌（hit）或 /21点 停牌（stand）")
+
+    def play_blackjack(self, account: dict[str, Any], action: str) -> str:
+        game = account.get("game")
+        if not game:
+            return "本人没有进行中的21点；/21点 开始 [金额]（/blackjack start）"
+        if action == "status":
+            return self.show_hand(game)
+        if action == "end":
+            index = int(game["index"])
+            bet = int(game["bet"])
+            append_game_record(
+                account,
+                kind="blackjack",
+                index=index,
+                outcome="主动认输",
+                stake=bet,
+                payout=0,
+                summary="主动结束牌局，不退下注",
+            )
+            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']}"
+        game["updated_at"] = time.time()
+        if action == "hit":
+            game["player"].append(game["deck"].pop())
+            if hand_value(game["player"]) > 21:
+                return self.finish_blackjack(account)
+            return self.show_hand(game)
+        while hand_value(game["dealer"]) < 17:
+            game["dealer"].append(game["deck"].pop())
+        return self.finish_blackjack(account)
+
+    @staticmethod
+    def finish_blackjack(account: dict[str, Any], *, natural: bool = False) -> str:
+        game = account["game"]
+        player = hand_value(game["player"])
+        dealer = hand_value(game["dealer"])
+        bet = game["bet"]
+        if player > 21 or natural and dealer == 21 and player != 21:
+            outcome, payout, stat = "你输了", 0, "losses"
+        elif player == dealer:
+            outcome, payout, stat = "平局", bet, "ties"
+        elif natural and player == 21:
+            outcome, payout, stat = "自然21点获胜", bet * 5 // 2, "wins"
+        elif dealer > 21 or player > dealer:
+            outcome, payout, stat = "你赢了", bet * 2, "wins"
+        else:
+            outcome, payout, stat = "你输了", 0, "losses"
+        account["balance"] += payout
+        account[stat] += 1
+        append_game_record(
+            account,
+            kind="blackjack",
+            index=int(game["index"]),
+            outcome=outcome,
+            stake=int(bet),
+            payout=int(payout),
+            summary=(
+                f"玩家{player}点，庄家{dealer}点"
+            ),
+        )
+        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']}")

packages/jwebot-sdk/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "jwebot-sdk"
-version = "0.6.0a0+1015"
+version = "0.6.3a0+1015"
 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__ = "0.6.0a0+1015"
+__version__ = "0.6.3a0+1015"

pyproject.toml

@@ -1,5 +1,5 @@
 [tool.jwebot]
-release = "alpha 0.6.0.1015"
+release = "alpha 0.6.3.1015"
 production_deployments_enabled_by_default = false
 
 [tool.jwebot.workspace]

tests/test_capture_reconcile.py

@@ -0,0 +1,93 @@
+"""Lost final-draw events must not strand an isolated incoming message."""
+
+import unittest
+from unittest import mock
+
+from test_wechat_local_ocr import MODULE
+
+
+class ReconcileTests(unittest.TestCase):
+    def test_final_drawing_without_damage_is_captured_and_duplicates_are_suppressed(self):
+        class Monitor:
+            window_id = 42
+
+            def __enter__(self):
+                return self
+
+            def __exit__(self, *_):
+                pass
+
+            def wait_for_message_change(self):
+                return next(self.changes)
+
+            changes = iter((True, False, False))
+
+        buffer = MODULE.CaptureFrameBuffer(5)
+        outcomes = []
+        offer = buffer.offer
+
+        def observe(frame):
+            outcome = offer(frame)
+            outcomes.append(outcome)
+            if len(outcomes) == 3:
+                buffer.close()
+            return outcome
+
+        images = [MODULE.Image.new("RGB", (2, 2), color) for color in ("white", "white", "black")]
+        recorder = MODULE.Recorder.__new__(MODULE.Recorder)
+        with mock.patch.object(MODULE, "XDamageMonitor", return_value=Monitor()), \
+                mock.patch.object(MODULE, "window_id", return_value="42"), \
+                mock.patch.object(MODULE, "capture_image", side_effect=images) as captures, \
+                mock.patch.object(MODULE, "message_view_at_loaded_bottom", return_value=True), \
+                mock.patch.object(MODULE, "message_image", side_effect=lambda image: image), \
+                mock.patch.object(MODULE, "incoming_frame_hashes", side_effect=[
+                    ("old-layout", "old-visual"), ("old-layout", "old-visual"), ("new-layout", "new-visual")]), \
+                mock.patch.object(buffer, "offer", side_effect=observe), \
+                mock.patch.object(MODULE, "telemetry_event") as telemetry:
+            recorder.capture_damage_frames(buffer)
+        self.assertEqual(captures.call_count, 3)
+        self.assertEqual(outcomes, ["enqueued", "duplicate", "enqueued"])
+        self.assertEqual(buffer.get().visual_hash, "old-visual")
+        self.assertEqual(buffer.get().visual_hash, "new-visual")
+        self.assertEqual([call.kwargs["source"] for call in telemetry.call_args_list
+                          if call.args == ("ocr.capture",)], ["damage", "reconcile", "reconcile"])
+        self.assertLessEqual(MODULE.DAMAGE_HEARTBEAT_SECONDS, 2)
+
+    def test_reconciliation_does_not_bypass_window_identity_check(self):
+        class Monitor:
+            window_id = 42
+
+            def __enter__(self):
+                return self
+
+            def __exit__(self, *_):
+                pass
+
+            def wait_for_message_change(self):
+                return False
+
+        buffer = MODULE.CaptureFrameBuffer(2)
+        recorder = MODULE.Recorder.__new__(MODULE.Recorder)
+        with mock.patch.object(MODULE, "XDamageMonitor", return_value=Monitor()), \
+                mock.patch.object(MODULE, "window_id", return_value="43"), \
+                mock.patch.object(MODULE, "capture_image") as capture:
+            recorder.capture_damage_frames(buffer)
+        capture.assert_not_called()
+        with self.assertRaises(MODULE.CaptureProducerError):
+            buffer.get()
+
+    def test_xlib_buffered_event_is_drained_before_waiting_for_socket(self):
+        monitor = MODULE.XDamageMonitor.__new__(MODULE.XDamageMonitor)
+        monitor._display = object()
+        monitor._descriptor = 42
+        monitor._x11 = mock.Mock()
+        monitor._x11.XPending.side_effect = [1, 0]
+        with mock.patch.object(monitor, "_drain_events", return_value=True) as drain, \
+                mock.patch.object(MODULE.select, "select", return_value=([], [], [])) as select:
+            self.assertTrue(monitor.wait_for_message_change())
+        drain.assert_called_once()
+        select.assert_called_once()
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_jwebot_edge.py

@@ -704,7 +704,7 @@ class LedgerAndApplicationTests(unittest.TestCase):
         self.assertEqual(response["route"], "local")
         self.assertEqual(response["execution_path"], "local-command")
         self.assertIn("由内核和插件描述自动生成", response["text"])
-        self.assertIn("[jwebot/dice]", response["text"])
+        self.assertIn("/丢骰子", response["text"])
         self.assertIn("骰子和多人骰子挑战", response["text"])
         self.assertIn("/steam", response["text"])
         self.assertIn("[DeepSeek]", response["text"])
@@ -1736,7 +1736,7 @@ class DirectClientTests(unittest.TestCase):
                 )
             )
             response = mock.MagicMock()
-            response.read.return_value = (
+            response.read1.return_value = (
                 'data: {"choices":[{"delta":{"content":"正"}}]}\n\n'
                 'data: {"choices":[{"delta":{"content":"常"}}]}\n\n'
                 "data: [DONE]\n\n"
@@ -1782,6 +1782,55 @@ class DirectClientTests(unittest.TestCase):
         finally:
             temporary.cleanup()
 
+    def test_stream_reader_stops_at_done_without_waiting_for_eof(self):
+        response = mock.MagicMock()
+        response.read1.side_effect = [
+            b'data: {"choices":[{"delta":{"content":"ok"}}]}\n'
+            b"data: [DO",
+            b"NE]\n\n",
+            AssertionError("reader continued after the stream completed"),
+        ]
+
+        raw = MODULE.read_direct_api_response(response)
+
+        self.assertEqual(
+            raw,
+            b'data: {"choices":[{"delta":{"content":"ok"}}]}\n'
+            b"data: [DONE]\n\n",
+        )
+        self.assertEqual(response.read1.call_count, 2)
+
+    def test_stream_reader_does_not_treat_content_as_done_marker(self):
+        response = mock.MagicMock()
+        response.read1.side_effect = [
+            b'data: {"choices":[{"delta":{"content":"data: [DONE]"}}]}\n\n',
+            b"data: [DONE]\n\n",
+            AssertionError("reader continued after the stream completed"),
+        ]
+
+        raw = MODULE.read_direct_api_response(response)
+
+        self.assertIn(b'"content":"data: [DONE]"', raw)
+        self.assertTrue(raw.endswith(b"data: [DONE]\n\n"))
+        self.assertEqual(response.read1.call_count, 2)
+
+    def test_stream_reader_enforces_total_wall_time(self):
+        response = mock.MagicMock()
+        response.read1.return_value = b": keep-alive\n\n"
+        with mock.patch.object(
+            MODULE.time,
+            "monotonic",
+            side_effect=[0.0, 0.0, 31.0],
+        ):
+            with self.assertRaisesRegex(
+                MODULE.DeepSeekUnavailable,
+                "exceeded total timeout",
+            ):
+                MODULE.read_direct_api_response(response)
+
+        response.fp.raw._sock.settimeout.assert_called_once_with(18.0)
+        response.read1.assert_called_once()
+
     def test_stream_parser_accepts_a_non_streaming_compatibility_payload(self):
         self.assertEqual(
             MODULE.deepseek_response_content(

tests/test_official_extension_inventory.py

@@ -69,7 +69,6 @@ class LocalExtensionInventoryTests(unittest.TestCase):
             with self.subTest(extension=extension.extension_id):
                 self.assertTrue(extension.description)
                 self.assertTrue(extension.help_entries)
-                self.assertIn(extension.extension_id, help_text)
                 self.assertIn(extension.description, help_text)
                 for entry in extension.help_entries:
                     self.assertIn(entry, help_text)
@@ -81,6 +80,9 @@ class LocalExtensionInventoryTests(unittest.TestCase):
                 for extension in LOCAL_EXTENSIONS
             },
             {
+                "context-summary": OFFICIAL_AUTHOR,
+                "person-directory": OFFICIAL_AUTHOR,
+                "single-player-games": JWEBOT_AUTHOR,
                 "person-memory": OFFICIAL_AUTHOR,
                 "server-status": OFFICIAL_AUTHOR,
                 "jwebot-version": OFFICIAL_AUTHOR,
@@ -108,6 +110,15 @@ class LocalExtensionInventoryTests(unittest.TestCase):
         self.assertEqual(
             combined,
             {
+                "single-player-games": (
+                    "wallet",
+                    "prize_draw",
+                    "lottery",
+                    "blackjack",
+                    "game_history",
+                    "game_rules",
+                ),
+                "person-directory": ("person_directory", "person_info"),
                 "person-memory": (
                     "remember",
                     "my_memory",

tests/test_release_061.py

@@ -0,0 +1,208 @@
+"""Behavioral coverage for local context, identity and single-player funds."""
+
+import json
+import random
+import tempfile
+import threading
+import unittest
+from pathlib import Path
+from unittest import mock
+
+from test_jwebot_edge import MODULE, config
+from jwebot_official_extensions.local_commands import local_command_parts
+from jwebot_official_extensions.player_games import hand_value
+
+
+class Release061Tests(unittest.TestCase):
+    def setUp(self):
+        self.temp = tempfile.TemporaryDirectory()
+        self.root = Path(self.temp.name)
+        self.engine = MODULE.EdgeApplication(config(self.root))
+        self.local = self.engine.local
+        self.counter = 0
+
+    def tearDown(self):
+        self.engine.ledger.connection.close()
+        self.temp.cleanup()
+
+    def command(self, question, *, sender="Alice", group="group-a", event_id=None):
+        self.counter += 1
+        row = {"question": question, "sender": sender, "group_fingerprint": group,
+               "event_id": event_id or f"test-{self.counter}"}
+        return self.local.handle(row, {})
+
+    def state(self, sender="Alice", group="group-a"):
+        key = self.engine.ledger.resolve_person(group, sender)["person_key"]
+        row = self.engine.ledger.connection.execute(
+            "SELECT state_json FROM player_accounts WHERE group_fingerprint=? AND person_key=?",
+            (group, key)).fetchone()
+        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"),
+                 ("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:
+            self.assertEqual(local_command_parts("/" + chinese), local_command_parts("/" + english))
+        self.assertEqual(local_command_parts("/Context200"), ("context_summary", "200"))
+
+    def test_context_scope_count_order_and_current_request_exclusion(self):
+        ledger = self.engine.ledger
+        for i in range(305):
+            ledger.record_repeat_history(event_id=f"history-{i:04d}", group_fingerprint="group-a",
+                                         sender="Alice", text=f"项目计划讨论编号 {i}",
+                                         observed_at=f"2026-10-15T00:{i // 60:02d}:{i % 60:02d}+00:00")
+        ledger.record_repeat_history(event_id="private-other", group_fingerprint="group-b",
+                                     sender="Bob", text="隔离群秘密话题", observed_at="2026-10-15T01:00:00+00:00")
+        for count in (100, 200, 300):
+            rows = ledger.recent_context_messages("group-a", count, exclude_event_id="history-0304")
+            self.assertEqual(len(rows), count)
+            self.assertEqual(rows[-1]["event_id"], "history-0303")
+            self.assertLess(rows[0]["observed_at"], rows[-1]["observed_at"])
+            with mock.patch.object(ledger, "bootstrap_repeat_history"):
+                result = self.command(f"/context {count}")
+            self.assertIn(f"实际{count}条", result)
+            self.assertNotIn("隔离群", result)
+        self.assertIn("用法", self.command("/context 101"))
+
+    def test_person_lookup_uses_alias_but_does_not_expose_steam_ids(self):
+        self.assertIn("alice", self.command("/people"))
+        self.assertIn("Alice is a configured test profile", self.command("/person a1"))
+        self.assertIn("群内索引", self.command("/个人属性"))
+        self.assertIn("没有", self.command("/person unconfigured"))
+        self.assertIn("只有", self.command("/people 99"))
+
+    def test_context_keeps_short_messages_omitted_from_repeat_statistics(self):
+        ledger = self.engine.ledger
+        self.assertFalse(ledger.record_repeat_history(
+            event_id="short", group_fingerprint="group-a", sender="Alice",
+            text="好", observed_at="2026-10-15T00:00:00+00:00"))
+        records = ledger.recent_context_messages("group-a", 100)
+        self.assertEqual([row["text"] for row in records], ["好"])
+        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"))
+        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("金币 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"))
+
+    def test_concurrent_replay_debits_once_and_rolls_back_failure(self):
+        replies = []
+        with mock.patch.object(self.local.players.rng, "randrange", return_value=99):
+            threads = [threading.Thread(target=lambda: replies.append(
+                self.command("/draw 100", event_id="concurrent"))) for _ in range(8)]
+            for thread in threads:
+                thread.start()
+            for thread in threads:
+                thread.join()
+        self.assertEqual(len(replies), 8)
+        self.assertEqual(len(set(replies)), 1)
+        self.assertEqual(self.state()["balance"], 900)
+        def broken(state):
+            state["balance"] = -1
+            return "bad"
+        with self.assertRaises(MODULE.LocalCommandError):
+            self.engine.ledger.apply_player_action("group-a", "Alice", "broken", "test", broken)
+        self.assertEqual(self.state()["balance"], 900)
+        with self.assertRaises(MODULE.LocalCommandError):
+            self.command("/draw 100", sender="Bob", event_id="concurrent")
+
+    def test_invalid_bets_insufficient_funds_and_lottery(self):
+        for question in ("/draw -1", "/draw 0", "/draw 101", "/lottery xyz",
+                         "/blackjack start -1", "/blackjack hit 10"):
+            self.assertRegex(self.command(question), "用法|下注")
+        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.assertEqual(self.state()["balance"], 0)
+
+    def test_blackjack_is_private_persistent_and_settles_exactly_once(self):
+        # Player 10+7, dealer 9+8: standing produces a tie.
+        def deck(cards):
+            ordered = [9, 6, 8, 7] + [card for card in cards if card not in {9, 6, 8, 7}]
+            cards[:] = ordered
+        with mock.patch.object(self.local.players.rng, "shuffle", side_effect=deck):
+            start = self.command("/21点 开始 50")
+        self.assertIn("暗牌", start)
+        self.assertEqual(self.state()["balance"], 950)
+        self.assertIn("没有进行中", self.command("/blackjack status", sender="Bob"))
+        self.assertIn("已有一局", self.command("/draw"))
+        self.engine.ledger.connection.close()
+        self.engine = MODULE.EdgeApplication(config(self.root))
+        self.local = self.engine.local
+        settled = self.command("/blackjack stand", sender="a1", event_id="stand-1")
+        self.assertIn("平局", settled)
+        self.assertEqual(self.state()["balance"], 1000)
+        self.assertEqual(self.command("/blackjack stand", event_id="stand-1"), settled)
+        self.assertIsNone(self.state()["game"])
+        self.assertEqual(self.state()["ties"], 1)
+
+    def test_aces_natural_bust_timeout_and_end(self):
+        self.assertEqual(hand_value([0, 13, 9]), 12)
+        self.assertEqual(hand_value([0, 9]), 21)
+        def natural(cards):
+            first = [0, 9, 8, 7]
+            cards[:] = first + [card for card in cards if card not in first]
+        with mock.patch.object(self.local.players.rng, "shuffle", side_effect=natural):
+            reply = self.command("/blackjack start 50")
+        self.assertIn("自然21点获胜", reply)
+        self.assertEqual(self.state()["balance"], 1075)
+        self.local.players.rng = random.Random(4)
+        self.command("/blackjack start 50")
+        if self.state()["game"]:
+            with mock.patch("jwebot_official_extensions.player_games.time.time", return_value=10**10):
+                self.assertIn("超时", self.command("/wallet"))
+            self.assertIsNone(self.state()["game"])
+        self.local.players.rng = random.Random(4)
+        self.command("/blackjack start 50")
+        if self.state()["game"]:
+            self.assertIn("不退下注", self.command("/blackjack end"))
+        self.assertIsNone(self.state()["game"])
+
+    def test_adapter_and_direct_paths_agree_for_new_plugins(self):
+        for index, question in enumerate(("/games", "/wallet", "/people", "/person a1", "/context 200")):
+            kind, _ = local_command_parts(question)
+            row = {"question": question, "sender": "Alice", "group_fingerprint": "group-a",
+                   "event_id": f"adapter-{index}"}
+            with mock.patch.object(self.engine.ledger, "bootstrap_repeat_history"):
+                direct = self.local.handle(row, {})
+                adapter = self.engine.extensions.handle(row, {})
+            self.assertEqual(adapter, direct)
+
+    def test_blackjack_bust_cannot_be_acted_on_by_another_player(self):
+        def bust_deck(cards):
+            first = [9, 22, 8, 7]
+            cards[:] = first + [card for card in cards if card not in first and card != 35] + [35]
+        with mock.patch.object(self.local.players.rng, "shuffle", side_effect=bust_deck):
+            self.command("/blackjack start 100")
+        self.assertIn("没有进行中", self.command("/blackjack hit", sender="Bob"))
+        self.assertEqual(len(self.state()["game"]["player"]), 2)
+        result = self.command("/blackjack hit", event_id="bust")
+        self.assertIn("你输了", result)
+        self.assertEqual(self.state()["balance"], 900)
+        self.assertIsNone(self.state()["game"])
+        self.assertEqual(self.command("/blackjack hit", event_id="bust"), result)
+        self.assertEqual(self.state()["losses"], 1)
+
+    def test_simultaneous_natural_blackjack_returns_stake(self):
+        def tied_deck(cards):
+            first = [0, 9, 13, 22]
+            cards[:] = first + [card for card in cards if card not in first]
+        with mock.patch.object(self.local.players.rng, "shuffle", side_effect=tied_deck):
+            self.assertIn("平局", self.command("/blackjack start 50"))
+        self.assertEqual(self.state()["balance"], 1000)
+        self.assertIsNone(self.state()["game"])
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_release_062.py

@@ -0,0 +1,128 @@
+"""Behavioral coverage for the alpha 0.6.2 latency and blackjack fixes."""
+
+import json
+import tempfile
+import unittest
+from pathlib import Path
+from unittest import mock
+
+from test_jwebot_edge import MODULE, config
+
+
+class Release062BlackjackTests(unittest.TestCase):
+    def setUp(self):
+        self.temp = tempfile.TemporaryDirectory()
+        self.engine = MODULE.EdgeApplication(config(Path(self.temp.name)))
+        self.counter = 0
+
+    def tearDown(self):
+        self.engine.ledger.connection.close()
+        self.temp.cleanup()
+
+    def command(self, question, *, event_id=None, sender="Alice"):
+        self.counter += 1
+        row = {
+            "question": question,
+            "sender": sender,
+            "group_fingerprint": "group-a",
+            "event_id": event_id or f"test-{self.counter}",
+        }
+        return self.engine.local.handle(row, {})
+
+    def state(self, sender="Alice"):
+        key = self.engine.ledger.resolve_person("group-a", sender)["person_key"]
+        row = self.engine.ledger.connection.execute(
+            """
+            SELECT state_json FROM player_accounts
+            WHERE group_fingerprint = ? AND person_key = ?
+            """,
+            ("group-a", key),
+        ).fetchone()
+        return json.loads(row[0])
+
+    @staticmethod
+    def hit_deck(cards):
+        first = [4, 5, 8, 7]
+        cards[:] = (
+            first
+            + [card for card in cards if card not in first and card != 1]
+            + [1]
+        )
+
+    def test_compact_start_sets_stake_then_separate_hit_replays_once(self):
+        with mock.patch.object(
+            self.engine.local.players.rng,
+            "shuffle",
+            side_effect=self.hit_deck,
+        ):
+            start = self.command("/21点开始100", event_id="compact-start")
+        self.assertTrue(start.startswith("alice #1："))
+        self.assertIn("下注 100", start)
+        self.assertEqual(self.state()["balance"], 900)
+        self.assertEqual(len(self.state()["game"]["player"]), 2)
+
+        first = self.command("/21点要牌", event_id="separate-hit")
+        replay = self.command(
+            "/21点 要牌",
+            event_id="separate-hit",
+            sender="a1",
+        )
+
+        self.assertEqual(replay, first)
+        self.assertTrue(first.startswith("alice #1："))
+        self.assertIn("13点", first)
+        self.assertEqual(self.state()["balance"], 900)
+        self.assertEqual(len(self.state()["game"]["player"]), 3)
+
+    def test_spaced_chinese_and_compact_english_start_use_bet(self):
+        def tied_deck(cards):
+            first = [9, 6, 8, 7]
+            cards[:] = first + [card for card in cards if card not in first]
+
+        for command, sender in (
+            ("/21点 开始 50", "Alice"),
+            ("/blackjackstart50", "Bob"),
+        ):
+            with self.subTest(command=command):
+                with mock.patch.object(
+                    self.engine.local.players.rng,
+                    "shuffle",
+                    side_effect=tied_deck,
+                ):
+                    started = self.command(command, sender=sender)
+                self.assertTrue(
+                    started.startswith(
+                        "alice #1：" if sender == "Alice" else "bob #1："
+                    )
+                )
+                self.assertIn("下注 50", started)
+                self.assertEqual(self.state(sender)["balance"], 950)
+                self.assertIsNotNone(self.state(sender)["game"])
+                self.command("/21点 结束", sender=sender)
+                self.assertIsNone(self.state(sender)["game"])
+
+    def test_default_bet_can_start_without_spaces(self):
+        with mock.patch.object(
+            self.engine.local.players.rng,
+            "shuffle",
+            side_effect=self.hit_deck,
+        ):
+            reply = self.command("/21点开始")
+
+        self.assertTrue(reply.startswith("alice #1："))
+        self.assertIn("下注 50", reply)
+        self.assertEqual(self.state()["balance"], 950)
+
+    def test_combined_start_and_action_explains_two_step_flow(self):
+        reply = self.command("/21点开始100要牌")
+
+        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")
+        self.assertIn("金币 1000", wallet)
+        self.assertIn("无进行中的牌局", wallet)
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_release_063.py

@@ -0,0 +1,252 @@
+"""Behavioral coverage for multiplayer lottery and indexed game records."""
+
+import json
+import tempfile
+import threading
+import unittest
+from pathlib import Path
+from unittest import mock
+
+from test_jwebot_edge import MODULE, config
+
+
+class Release063GameTests(unittest.TestCase):
+    def setUp(self):
+        self.temp = tempfile.TemporaryDirectory()
+        self.engine = MODULE.EdgeApplication(config(Path(self.temp.name)))
+        self.counter = 0
+
+    def tearDown(self):
+        self.engine.ledger.connection.close()
+        self.temp.cleanup()
+
+    def command(
+        self,
+        question,
+        *,
+        sender="Alice",
+        group="group-a",
+        event_id=None,
+    ):
+        self.counter += 1
+        row = {
+            "question": question,
+            "sender": sender,
+            "group_fingerprint": group,
+            "event_id": event_id or f"test-{self.counter}",
+        }
+        return self.engine.local.handle(row, {})
+
+    def state(self, sender="Alice", group="group-a"):
+        key = self.engine.ledger.resolve_person(group, sender)["person_key"]
+        row = self.engine.ledger.connection.execute(
+            """
+            SELECT state_json FROM player_accounts
+            WHERE group_fingerprint = ? AND person_key = ?
+            """,
+            (group, key),
+        ).fetchone()
+        return json.loads(row[0])
+
+    def test_lottery_number_forms_sort_and_validate(self):
+        parse = self.engine.local.players.parse_lottery_entry
+
+        self.assertEqual(parse("347"), ((3, 4, 7), 10))
+        self.assertEqual(parse("743"), ((3, 4, 7), 10))
+        self.assertEqual(parse("报名 3 4 7 100"), ((3, 4, 7), 100))
+        self.assertEqual(parse("报名347100"), ((3, 4, 7), 100))
+        self.assertEqual(parse("743金额500"), ((3, 4, 7), 500))
+        with self.assertRaisesRegex(ValueError, "互不重复"):
+            parse("337")
+        with self.assertRaisesRegex(ValueError, "1～500"):
+            parse("347501")
+
+    def test_multiplayer_lottery_status_draw_and_payouts(self):
+        self.assertIn("第 1 把已开始", self.command("/彩票开始"))
+        self.assertIn("alice #1", self.command("/彩票347100"))
+        self.assertIn(
+            "bob #1",
+            self.command("/彩票 报名 9 3 7 200", sender="Bob"),
+        )
+        self.assertIn(
+            "carol #1",
+            self.command("/彩票 1 2 3 100", sender="Carol"),
+        )
+        self.assertIn(
+            "Dave #1",
+            self.command("/彩票045100", sender="Dave"),
+        )
+
+        status = self.command("/彩票当前状态", sender="Bob")
+        self.assertIn("已报名 4 人", status)
+        self.assertIn("alice #1：3 4 7，下注 100", status)
+        self.assertIn("bob #1：3 7 9，下注 200", status)
+        self.assertIn("只有本轮主持人 alice", self.command(
+            "/彩票开奖", sender="Bob"))
+
+        with mock.patch.object(
+            self.engine.local.players.rng,
+            "sample",
+            return_value=[7, 3, 9],
+        ):
+            result = self.command("/彩票开奖", event_id="draw-round-1")
+
+        self.assertIn("开奖号：3 7 9", result)
+        self.assertIn("alice #1：命中 2 个", result)
+        self.assertIn("返还 120", result)
+        self.assertIn("bob #1：命中 3 个", result)
+        self.assertIn("返还 600", result)
+        self.assertIn("carol #1：命中 1 个", result)
+        self.assertIn("返还 50", result)
+        self.assertIn("Dave #1：命中 0 个", result)
+        self.assertEqual(self.state()["balance"], 1020)
+        self.assertEqual(self.state("Bob")["balance"], 1400)
+        self.assertEqual(self.state("Carol")["balance"], 950)
+        self.assertEqual(self.state("Dave")["balance"], 900)
+        self.assertIn("命中2个", self.command("/游戏记录"))
+        self.assertIn("命中3个", self.command(
+            "/个人记录", sender="Bob"))
+        self.assertIn("没有进行中", self.command("/彩票当前状态"))
+
+    def test_registration_update_keeps_index_and_cancel_refunds(self):
+        self.command("/彩票 开始")
+        first = self.command("/彩票 347 100")
+        updated = self.command("/彩票 743 200", sender="a1")
+
+        self.assertIn("alice #1", first)
+        self.assertIn("alice #1", updated)
+        self.assertIn("报名已更新", updated)
+        self.assertEqual(self.state()["balance"], 800)
+        status = self.command("/彩票 状态")
+        self.assertIn("已报名 1 人", status)
+        self.assertIn("3 4 7，下注 200", status)
+        self.assertIn("只有本轮主持人 alice", self.command(
+            "/彩票 取消", sender="Bob"))
+        self.assertIn("已全额退回", self.command("/彩票取消"))
+        self.assertEqual(self.state()["balance"], 1000)
+        self.assertIn("净变化 +0", self.command("/游戏记录"))
+
+        self.command("/彩票开始")
+        second = self.command("/彩票347")
+        self.assertIn("alice #2", second)
+
+    def test_concurrent_registration_replay_debits_once(self):
+        self.command("/彩票开始")
+        replies = []
+        errors = []
+
+        def register():
+            try:
+                replies.append(
+                    self.command("/彩票347500", event_id="same-entry")
+                )
+            except BaseException as error:
+                errors.append(error)
+
+        threads = [threading.Thread(target=register) for _ in range(8)]
+        for thread in threads:
+            thread.start()
+        for thread in threads:
+            thread.join()
+
+        self.assertFalse(errors)
+        self.assertEqual(len(set(replies)), 1)
+        self.assertEqual(self.state()["balance"], 500)
+        self.assertIn("已报名 1 人", self.command("/彩票状态"))
+
+    def test_blackjack_uses_1_to_500_and_indexed_history(self):
+        def tied_deck(cards):
+            first = [9, 6, 8, 7]
+            cards[:] = first + [card for card in cards if card not in first]
+
+        with mock.patch.object(
+            self.engine.local.players.rng,
+            "shuffle",
+            side_effect=tied_deck,
+        ):
+            started = self.command("/21点开始500")
+        self.assertTrue(started.startswith("alice #1："))
+        self.assertEqual(self.state()["balance"], 500)
+        settled = self.command("/21点停牌")
+        self.assertTrue(settled.startswith("alice #1："))
+        self.assertIn("平局", settled)
+        self.assertEqual(self.state()["balance"], 1000)
+        history = self.command("/游戏记录")
+        self.assertIn("alice #1：21点 平局", history)
+        self.assertIn("下注 500，返还 500", history)
+        self.assertIn("1～500", self.command("/21点开始501"))
+
+    def test_prize_draw_has_indexed_history(self):
+        with mock.patch.object(
+            self.engine.local.players.rng,
+            "randrange",
+            return_value=99,
+        ):
+            first = self.command("/抽奖10")
+            second = self.command("/抽奖10")
+
+        self.assertTrue(first.startswith("alice #1："))
+        self.assertTrue(second.startswith("alice #2："))
+        history = self.command("/gamehistory")
+        self.assertIn("alice #2：抽奖 未中奖", history)
+        self.assertIn("alice #1：抽奖 未中奖", history)
+
+    def test_open_round_and_history_survive_restart(self):
+        root = Path(self.temp.name)
+        self.command("/彩票开始")
+        self.command("/彩票347100")
+        self.engine.ledger.connection.close()
+        self.engine = MODULE.EdgeApplication(config(root))
+
+        self.assertIn("alice #1：3 4 7", self.command("/彩票状态"))
+        with mock.patch.object(
+            self.engine.local.players.rng,
+            "sample",
+            return_value=[3, 4, 7],
+        ):
+            self.command("/彩票开奖")
+        self.engine.ledger.connection.close()
+        self.engine = MODULE.EdgeApplication(config(root))
+
+        history = self.command("/游戏记录")
+        self.assertIn("alice #1：彩票 命中3个", history)
+        self.assertEqual(self.state()["balance"], 1200)
+
+    def test_registration_rolls_back_when_response_building_fails(self):
+        ledger = self.engine.ledger
+        ledger.start_group_lottery(
+            "group-a",
+            "Alice",
+            "start",
+            "lottery:start",
+            lambda _: "started",
+        )
+
+        def fail(_):
+            raise RuntimeError("render failed")
+
+        with self.assertRaisesRegex(RuntimeError, "render failed"):
+            ledger.register_group_lottery(
+                "group-a",
+                "Alice",
+                "register",
+                "lottery:347100",
+                (3, 4, 7),
+                100,
+                fail,
+            )
+
+        account = ledger.connection.execute(
+            "SELECT COUNT(*) FROM player_accounts"
+        ).fetchone()[0]
+        entries = ledger.connection.execute(
+            "SELECT COUNT(*) FROM group_lottery_entries"
+        ).fetchone()[0]
+        actions = ledger.connection.execute(
+            "SELECT COUNT(*) FROM player_actions WHERE event_id = 'register'"
+        ).fetchone()[0]
+        self.assertEqual((account, entries, actions), (0, 0, 0))
+
+
+if __name__ == "__main__":
+    unittest.main()

wechat_local_ocr.py

@@ -63,7 +63,7 @@ POLL_SECONDS = 1.0
 STABLE_FRAMES = 2
 DAMAGE_DEBOUNCE_SECONDS = 0.35
 DAMAGE_MAX_BURST_SECONDS = 1.0
-DAMAGE_HEARTBEAT_SECONDS = 60.0
+DAMAGE_HEARTBEAT_SECONDS = 2.0
 CAPTURE_BUFFER_FRAMES = 12
 OCR_MESSAGE_CACHE_SIZE = 256
 OCR_CACHE_REFRESH_SCANS = 120
@@ -1060,7 +1060,11 @@ class XDamageMonitor:
             timeout = max(0.0, deadline - now)
             if timeout == 0:
                 return first_relevant is not None
-            readable, _, _ = select.select([self._descriptor], [], [], timeout)
+            # Xlib may already hold events even when its socket is no longer readable.
+            if self._x11.XPending(self._display):
+                readable = [self._descriptor]
+            else:
+                readable, _, _ = select.select([self._descriptor], [], [], timeout)
             if not readable:
                 return first_relevant is not None
             if not self._drain_events():
@@ -1728,10 +1732,8 @@ class Recorder:
                     buffer.capacity,
                 )
                 while not buffer.closed:
-                    if not monitor.wait_for_message_change():
+                    changed = monitor.wait_for_message_change()
-                        if int(window_id()) != monitor.window_id:
+                    source = "damage" if changed else "reconcile"
-                            raise RuntimeError("WeChat window changed")
-                        continue
                     identifier = window_id()
                     if int(identifier) != monitor.window_id:
                         raise RuntimeError("WeChat window changed")
@@ -1746,7 +1748,7 @@ class Recorder:
                         * 1000,
                         outcome="captured",
                         phase="initial_capture",
-                        source="damage",
+                        source=source,
                     )
                     latest_synced = message_view_at_loaded_bottom(image)
                     if not latest_synced:
@@ -1757,7 +1759,7 @@ class Recorder:
                             elapsed_ms=(time.monotonic() - follow_started) * 1000,
                             outcome="followed" if followed else "lock_busy",
                             phase="follow_latest",
-                            source="damage",
+                            source=source,
                         )
                         if followed:
                             recapture_started = time.monotonic()
@@ -1770,7 +1772,7 @@ class Recorder:
                                 * 1000,
                                 outcome="captured",
                                 phase="recapture",
-                                source="damage",
+                                source=source,
                             )
                             latest_synced = message_view_at_loaded_bottom(image)
                     crop = message_image(image)
@@ -1801,7 +1803,7 @@ class Recorder:
                         elapsed_ms=(time.monotonic() - capture_started) * 1000,
                         outcome=outcome,
                         depth=stats["depth"],
-                        source="damage",
+                        source=source,
                     )
                     if outcome == "closed":
                         return