## 更新内容
* fix(core): 在模式验证中强制转换字符串类型的布尔值 by @tt-a1i in https://github.com/QwenLM/qwen-code/pull/1284
* fix(cli): 当 disableUpdateNag 为 true 时跳过更新检查 by @liqiongyu in https://github.com/QwenLM/qwen-code/pull/1397
* feat(i18n): 新增德语语言支持 by @afarber in https://github.com/QwenLM/qwen-code/pull/1378
* Feat/javasdk by @gwinthis in https://github.com/QwenLM/qwen-code/pull/1412
* fix: 优化非交互模式下的工具执行反馈 by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/1383
* fix: 在文本模式下，API 错误时以非零代码退出 by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/1376
* fix: 为 stream-json 输出格式保留思考内容中的空白字符 by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/1365
* feat: 将 `--experimental-acp` 正式升级为稳定版 `--acp` 标志 by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/1355
* Doc/qwencode java by @gwinthis in https://github.com/QwenLM/qwen-code/pull/1414
* feat(cli): 为 /approval-mode 命令新增直接参数支持 by @tt-a1i in https://github.com/QwenLM/qwen-code/pull/1391
* fix: 改进 Windows 后台进程处理与清理 by @xuewenjie123 in https://github.com/QwenLM/qwen-code/pull/1146
* Fix resume command broken after new chat by @DennisYu07 in https://github.com/QwenLM/qwen-code/pull/1374
* chore: 版本升级至 0.6.1 by @Mingholy in https://github.com/QwenLM/qwen-code/pull/1423
* fix(cli,core): 在非交互运行中遵循 `tools.core` / `tools.allowed` by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/1406
* fix(core): 不对 OpenAI 兼容 API 强制设置 reasoning/topP 默认值 by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/1415
* [OpenaiContentGenerate] convertOpenAIResponseToGemini 记录 thoughtsTokenCount by @Weaxs in https://github.com/QwenLM/qwen-code/pull/1393
* feat: 优化提示“非好友…”的错误消息问题 by @BlockHand in https://github.com/QwenLM/qwen-code/pull/1282
* chore: 版本升级至 0.7.0 by @Mingholy in https://github.com/QwenLM/qwen-code/pull/1434
* fix(core): 确保在无头模式下始终显示 OAuth URL by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/1426
* feat: 修改用户设置与工作区设置的优先级顺序 by @BlockHand in https://github.com/QwenLM/qwen-code/pull/1433
* feat: 支持多提供商模型配置 by @Mingholy in https://github.com/QwenLM/qwen-code/pull/1291
* refactor: 将 IDE 上下文从 JSON 转换为纯文本格式 by @xuewenjie123 in https://github.com/QwenLM/qwen-code/pull/1424
* fix: 解决多提供商冷启动问题 by @Mingholy in https://github.com/QwenLM/qwen-code/pull/1439
* fix(cli): /memory show 命令现在尊重 context.fileName by @liqiongyu in https://github.com/QwenLM/qwen-code/pull/1428
* fix: 解决 macOS 和 Windows 上外部编辑器启动失败的问题 by @xuewenjie123 in https://github.com/QwenLM/qwen-code/pull/1351
* fix(core): 处理 OpenAI 流式分块中缺失的 delta 字段 by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/1448
* fix(cli): 在 Linux 上默认设置沙箱 UID/GID 映射 by @liqiongyu in https://github.com/QwenLM/qwen-code/pull/1453
* fix(shell): 防止 Windows 前台任务出现控制台窗口闪烁 by @xuewenjie123 in https://github.com/QwenLM/qwen-code/pull/1464
* fix(cli): 对已弃用或未知的设置键名发出警告 by @liqiongyu in https://github.com/QwenLM/qwen-code/pull/1427
* fix(core): 改进 OAuth fetch-failed 诊断信息 by @liqiongyu in https://github.com/QwenLM/qwen-code/pull/1457
* fix: 优化 SDK 发布工作流并提升稳定性 by @Mingholy in https://github.com/QwenLM/qwen-code/pull/1462
## 新贡献者
* @tt-a1i 在 https://github.com/QwenLM/qwen-code/pull/1284 完成了首次贡献
* @liqiongyu 在 https://github.com/QwenLM/qwen-code/pull/1397 完成了首次贡献
* @gwinthis 在 https://github.com/QwenLM/qwen-code/pull/1412 完成了首次贡献
* @LaZzyMan 在 https://github.com/QwenLM/qwen-code/pull/1383 完成了首次贡献
* @DennisYu07 在 https://github.com/QwenLM/qwen-code/pull/1374 完成了首次贡献
**完整更新日志**: https://github.com/QwenLM/qwen-code/compare/sdk-typescript-v0.1.1...sdk-typescript-v0.1.2
## What's Changed
* fix(core): coerce string boolean values in schema validation by @tt-a1i in https://github.com/QwenLM/qwen-code/pull/1284
* fix(cli): skip update check when disableUpdateNag is true by @liqiongyu in https://github.com/QwenLM/qwen-code/pull/1397
* feat(i18n): add German language support by @afarber in https://github.com/QwenLM/qwen-code/pull/1378
* Feat/javasdk by @gwinthis in https://github.com/QwenLM/qwen-code/pull/1412
* fix: improve tool execution feedback in non-interactive mode by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/1383
* fix: exit with non-zero code on API errors in text mode by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/1376
* fix: preserve whitespace in thinking content for stream-json output format by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/1365
* feat: graduate `--experimental-acp` to stable `--acp` flag by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/1355
* Doc/qwencode java by @gwinthis in https://github.com/QwenLM/qwen-code/pull/1414
* feat(cli): add direct argument support for /approval-mode command by @tt-a1i in https://github.com/QwenLM/qwen-code/pull/1391
* fix: improve windows background process handling and cleanup by @xuewenjie123 in https://github.com/QwenLM/qwen-code/pull/1146
* Fix resume command broken after new chat by @DennisYu07 in https://github.com/QwenLM/qwen-code/pull/1374
* chore: bump version to 0.6.1 by @Mingholy in https://github.com/QwenLM/qwen-code/pull/1423
* fix(cli,core): honor `tools.core` / `tools.allowed` in non-interactive runs by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/1406
* fix(core): don’t force reasoning/topP defaults for OpenAI-compatible APIs by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/1415
* [OpenaiContentGenerate] convertOpenAIResponseToGemini record thoughtsTokenCount by @Weaxs in https://github.com/QwenLM/qwen-code/pull/1393
* feat: Optimize the issue where an error message indicating unfriendli… by @BlockHand in https://github.com/QwenLM/qwen-code/pull/1282
* chore: bump version to 0.7.0 by @Mingholy in https://github.com/QwenLM/qwen-code/pull/1434
* fix(core): ensure OAuth URL always displayed in headless mode by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/1426
* feat: Modify the selection order of user Settings and workspace Settings by @BlockHand in https://github.com/QwenLM/qwen-code/pull/1433
* feat: multi-provider models config support by @Mingholy in https://github.com/QwenLM/qwen-code/pull/1291
* refactor: convert IDE context from JSON to plain text format by @xuewenjie123 in https://github.com/QwenLM/qwen-code/pull/1424
* fix: multi provider cold start issue by @Mingholy in https://github.com/QwenLM/qwen-code/pull/1439
* fix(cli): /memory show respects context.fileName by @liqiongyu in https://github.com/QwenLM/qwen-code/pull/1428
* fix: resolve external editor launch failure on macOS and Windows by @xuewenjie123 in https://github.com/QwenLM/qwen-code/pull/1351
* fix(core): handle missing delta in OpenAI stream chunks by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/1448
* fix(cli): default sandbox UID/GID mapping on Linux by @liqiongyu in https://github.com/QwenLM/qwen-code/pull/1453
* fix(shell): prevent console window flash on Windows for foreground tasks by @xuewenjie123 in https://github.com/QwenLM/qwen-code/pull/1464
* fix(cli): warn on deprecated/unknown settings keys by @liqiongyu in https://github.com/QwenLM/qwen-code/pull/1427
* fix(core): improve OAuth fetch-failed diagnostics by @liqiongyu in https://github.com/QwenLM/qwen-code/pull/1457
* fix: SDK release workflow and stability improvements by @Mingholy in https://github.com/QwenLM/qwen-code/pull/1462
## New Contributors
* @tt-a1i made their first contribution in https://github.com/QwenLM/qwen-code/pull/1284
* @liqiongyu made their first contribution in https://github.com/QwenLM/qwen-code/pull/1397
* @gwinthis made their first contribution in https://github.com/QwenLM/qwen-code/pull/1412
* @LaZzyMan made their first contribution in https://github.com/QwenLM/qwen-code/pull/1383
* @DennisYu07 made their first contribution in https://github.com/QwenLM/qwen-code/pull/1374
**Full Changelog**: https://github.com/QwenLM/qwen-code/compare/sdk-typescript-v0.1.1...sdk-typescript-v0.1.2