Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
聚焦生活陪伴场景的 Agent 聊天记录可视化工具,支持无感记录聊天内容,并将其展示为时间轴、统计分析视图,支持本地部署、自动化截图发送与持续分享
| Date | Stars |
|---|---|
| 2026-07-31 | 353 |
| 2026-08-06 | 353 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
<div align="center"> [中文](./README.md) · [English](./README.en.md) # Timeline for Agent 让 agent 在长期对话里自然沉淀用户的生活轨迹,并在需要时生成可筛选、可截图的时间轴报表。 当前先开放 CLI,MCP 能力正在开发中。 [](./LICENSE) [](./package.json) [](#agent-guide) [](#screenshots) <p> <a href="#preview">预览</a> · <a href="#user-guide">用户使用</a> · <a href="#agent-guide">Agent 接入</a> · <a href="#screenshots">截图规则</a> </p> </div> > 重要前提:agent 上下文里必须带有明确对话时间戳,否则写入的时间数据会不准。如果你手头还没有满足这个前提的 agent,可以直接使用我的赛博老板项目:[cyberboss](https://github.com/WenXiaoWendy/cyberboss)。 它适合这样的场景: - agent 在和用户的持续聊天中,会掌握用户的生活作息和身体状况,这些信息会不断累积,但没有一个结构化的出口 - 用户希望这些片段被稳定记录成生活轨迹时间轴,而不是散落在长对话里 - 当用户想看某一天、某一周、某个月,或某个分类、某个明细时,可以立刻从 agent 那里获取对应的报表截图 <a id="preview"></a> ## Dashboard 预览 桌面主视图示例:  移动端视图示例: | 时间轴 | 分析 | 事件 | | --- | --- | --- | |  |  |  | 局部筛选示例: 需要编码类别的分析局部截图时,可描述为“截 2026-04 的月视图分析区,筛到工作 > 编码”:  agent 发送截图给用户的示例:  其他局部截图和排列组合示例可在 `examples/` 查看: - `timeline-dashboard-timeline-view.png` - `timeline-dashboard-analytics-view.png` - `timeline-dashboard-events-view.png` - `screenshot-combos/timeline-dashboard-main-view-zh.png` - `screenshot-combos/timeline-dashboard-mobile-timeline-view.png` - `screenshot-combos/timeline-dashboard-mobile-analytics-view.png` - `screenshot-combos/timeline-dashboard-mobile-events-view.png` - `screenshot-combos/week-main-default.png` - `screenshot-combos/week-analytics-default.png` - `screenshot-combos/month-2026-04-work-coding-analytics.png` - `screenshot-combos/week-rest-sleep-events.png` - `screenshot-combos/day-2026-04-05-timeline.png` - `screenshot-combos/month-2026-04-analytics.png` <a id="user-guide"></a> ## 用户使用 ### 最常见的两种用法 1. 让 agent 在对话中持续写入时间轴,需要时再截图给用户 2. 自己在本地长期挂一个 dashboard,随时打开查看,不依赖 agent 临时截图 ### 如果想长期开着本地页面随时查看 如果你不想每次都依赖 agent 触发截图,而是希望自己在本地随时打开 dashboard 看当前数据,最直接的方式是长期运行下面两个命令中的一个: ```bash timeline-for-agent dev timeline-for-agent serve ``` 区别是: - `timeline-for-agent dev` 适合开发和频繁改数据,会监听源码和数据文件,自动重建并热刷新 - `timeline-for-agent serve` 适合稳定查看,直接把已经构建好的静态页面挂起来,不做监听 常见用法: ```bash timeline-for-agent build timeline-for-agent serve --port 4317 ``` 如果你长期只想“查看”而不是“边改边看”,优先用 `serve`。如果你会一边改数据一边观察效果,优先用 `dev`。 ### 安装与启动 环境前提: - Node.js `>=22` - 已执行 `npm install` - 如果需要截图,本机需要可用的 Chromium / Chrome / Edge,或可被 `playwright-core` 发现的浏览器 最快启动方式: ```bash npm install npm run timeline-dev ``` 默认会启动在 `http://127.0.0.1:4317`。 如果不想走 `npm scripts`,也可以直接用: ```bash node ./bin/timeline-for-agent.js dev ``` ### 数据放在哪里 默认状态目录在 `~/.timeline-for-agent/timeline/`,主要文件包括: - `timeline-state.json` 单一权威状态快照 - `timeline-taxonomy.json` 分类、子类和 eventNode - `timeline-facts.json` 真实时间轴事件数据 - `site/` `build` 产出的静态页面 - `shots/` `screenshot` 默认输出目录 时区配置: - `timeline-state.json`、`timeline-taxonomy.json`、`timeline-facts.json` 都会带 `timezone` - dashboard 的日/周/月分桶、趋势统计、tooltip 时间、移动端时间轴定位,都应以这个 `timezone` 为准 - 如果你部署在东八区外,或希望按其他时区解释同一份事件数据,优先检查并修改 `timeline-state.json` 里的 `timezone` - 如果你还在使用拆分文件模式,也需要保证 `timeline-taxonomy.json` 和 `timeline-facts.json` 里的 `timezone` 一致 CLI 会优先读取这两个 `.env` 位置: - 当前工作目录下的 `.env` - `~/.timeline-for-agent/.env` 常用环境变量: - `TIMELINE_FOR_AGENT_LOCALE` - `TIMELINE_FOR_AGENT_STATE_DIR` - `TIMELINE_FOR_AGENT_DIR` - `TIMELINE_FOR_AGENT_PORT`
Excerpt of 8,337 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:6ad79bc4e5b88e05, llm:description: Chinese — 'Agent chat record visualization tool focused on companionship scenarios; supports seamless recording of chat content and displays as timeline and analytics views; supports local deployment, automated screenshot sending and continuous sharing'
matched fp:6ad79bc4e5b88e05, llm:description: Chinese — 'Agent chat record visualization tool focused on companionship scenarios; supports seamless recording of chat content and displays as timeline and analytics views; supports local deployment, automated screenshot sending and continuous sharing'
matched fp:6ad79bc4e5b88e05, llm:description: Chinese — 'Agent chat record visualization tool focused on companionship scenarios; supports seamless recording of chat content and displays as timeline and analytics views; supports local deployment, automated screenshot sending and continuous sharing'