{Full Name}【免费下载链接】rowboatOpen-source AI coworker, with memory项目地址: https://gitcode.com/GitHub_Trending/rowb/rowboatInfoRole:{role, or inferred role with qualifier, or leave blank if truly unknown}Organization:[[Organizations/{organization}]] or leave blankEmail:{email or leave blank}Aliases:{comma-separated: first name, nicknames, email}First met:{YYYY-MM-DD}Last update:{YYYY-MM-DD}Summary{2-3 sentences: Who they are, whether I know them through an ongoing relationship or met them in a specific encounter, and what were discussing or working on together...}Activity{YYYY-MM-DD}({meeting|email|voice memo}): {Summary with [[Folder/Name]] links}Key factsOpen itemsAssistant notes注意模板中的 [[Organizations/{organization}]] 这类双括号链接——笔记之间通过双向链接形成一张**知识图谱**这是实体化的直接体现。模板还预留了 Assistant notes 段供 AI 写入面向未来的助手侧观察由 [note_system.ts](https://link.gitcode.com/i/020c855dd458eab87a2fff7c1291ca5b) 的 renderTemplateWithAssistantNotes() 在渲染时自动补齐。 ### 2.2 自动更新的上下文Auto-updating context 欢迎文档承诺当新的邮件和会议到来时Rowboat 会把决策decisions、承诺commitments和相关上下文追加到恰当的笔记中。这条流水线的真实执行路径是 1. **同步Sync**[sync_gmail.ts](https://link.gitcode.com/i/e6a65e39e5ef7c8dfd23fb28bb253bf0) 等同步脚本从 Gmail / Fireflies / Outlook 拉取新数据写入 WorkDir 下对应目录如 gmail_sync/、calendar_sync/并在成功后触发知识图谱构建。 2. **变更检测Change detection**[graph_state.ts](https://link.gitcode.com/i/4d48105508eea86636442f1979e9a54f) 采用 **mtime 内容哈希的混合策略**先快速比对文件修改时间mtime 未变则直接跳过若 mtime 变化再计算 SHA-256 内容哈希确认是否真的改变从而避免touch 一下就被重复处理的误报。处理状态记录在 WorkDir/knowledge_graph_state.json json { processedFiles: { /path/to/file.md: { mtime: 2026-01-07T10:30:00.000Z, hash: a3f5e9d2c8b1..., lastProcessed: 2026-01-07T10:35:00.000Z } }, lastBuildTime: 2026-01-07T10:35:00.000Z }【免费下载链接】rowboatOpen-source AI coworker, with memory项目地址: https://gitcode.com/GitHub_Trending/rowb/rowboat创作声明：本文部分内容由AI辅助生成（AIGC），仅供参考