Wienerdog – memory and self-improving skills for Claude Code/Codex 是一个「skill」类 AI agent,解决:在编程过程中记住用户习惯与历史上下文，持续优化代码建议与自动补全。定价:产品页未标明。截至 2026-10-27,KanonAgent 记录到 8 票。本站首次收录于 2026-08-01。
为 Claude Code 和 Codex 提供记忆与自我进化能力的 AI 助手，能记住过往交互、持续优化编程建议，提升开发效率
8 票
Kanon 于 2026 年 8 月 1 日 收录
状态依据:npm_downloads_7d momentum fell to 0 per 3d from 21 (<= 40% of the prior window)
📈 时间线什么时候变了什么
- 降温 npm_downloads_7d momentum fell to 0 per 3d from 21 (<= 40% of the prior window)
- 降温 npm_downloads_7d momentum fell to -1 per 3d from 17 (<= 40% of the prior window)
- 降温 npm_downloads_7d momentum fell to -1 per 3d from 17 (<= 40% of the prior window)
- 降温 npm_downloads_7d momentum fell to -176 per 3d from 18 (<= 40% of the prior window)
- 降温 npm_downloads_7d momentum fell to -176 per 3d from 14 (<= 40% of the prior window)
- 降温 npm_downloads_7d momentum fell to -139 per 3d from 15 (<= 40% of the prior window)
- 降温 npm_downloads_7d momentum fell to -1 per 3d from 54 (<= 40% of the prior window)
- 降温 npm_downloads_7d momentum fell to 14 per 3d from 157 (<= 40% of the prior window)
- 重启 npm_downloads_7d did not move for 7d, then gained 14 in the last 3d
- 重启 npm_downloads_7d did not move for 7d, then gained 14 in the last 3d
- 降温 npm_downloads_7d momentum fell to 14 per 3d from 157 (<= 40% of the prior window)
- 加速 npm_downloads_7d gained 15 in the last 3d after a flat 3d baseline (no prior movement to compare a ratio against)
- 异常 daily change -158.0 is -7.0 standard deviations from this entity's own prior mean (7.8±23.7, baseline excludes this point) — unusual, not necessarily adoption
- 异常 daily change -158.0 is -7.0 standard deviations from this entity's own prior mean (7.8±23.7, baseline excludes this point) — unusual, not necessarily adoption
- 降温 npm_downloads_7d momentum fell to -14 per 3d from 168 (<= 40% of the prior window)
- 降温 npm_downloads_7d momentum fell to -37 per 3d from 239 (<= 40% of the prior window)
- 降温 npm_downloads_7d momentum fell to -12 per 3d from 214 (<= 40% of the prior window)
- 加速 npm_downloads_7d gained 168 in the last 3d vs 48 in the 3d before (ratio 3.5x, threshold 1.5x)
- 降温 npm_downloads_7d momentum fell to -12 per 3d from 214 (<= 40% of the prior window)
- 加速 npm_downloads_7d gained 239 in the last 3d after a flat 3d baseline (no prior movement to compare a ratio against)
- 加速 npm_downloads_7d gained 214 in the last 3d after a flat 3d baseline (no prior movement to compare a ratio against)
- 异常 daily change 83.0 is 4.0 standard deviations from this entity's own prior mean (6.8±19.0, baseline excludes this point) — unusual, not necessarily adoption
- 加速 npm_downloads_7d gained 48 in the last 3d after a flat 3d baseline (no prior movement to compare a ratio against)
- 异常 daily change 83.0 is 4.0 standard deviations from this entity's own prior mean (6.8±19.0, baseline excludes this point) — unusual, not necessarily adoption
🔎 已知 / 未知
23 个字段里 12 个未知
| 可机器调用 | — | 未知 | — |
| 开源 | 1 | 推断 | 1 条引文 |
| 可自托管 | 1 | 推断 | 1 条引文 |
| 需自带 key | — | 未知 | — |
| 自主度 | 2 | 推断 | 3 条引文 |
| 计费模式 | — | 未知 | — |
| 集成面 | — | 未知 | — |
「未知」= 我们没核实过,**不等于「否」**。硬性筛选永远不会把未知当成否。
🤖 Agent 拆解 · skill
解决什么场景在编程过程中记住用户习惯与历史上下文，持续优化代码建议与自动补全
自主度L2 · 有工具调用(引文: "a nightly "dreaming" process that learns from your conversat…")
前置条件开源 · 可自托管
为什么值得关注
突破传统 AI 编程助手无记忆的局限，实现长期学习与上下文延续，是 AI agent 的典型应用
证据引文判定所依据的原文 · 逐字引用
“public GitHub repository (README fetched)”— structural
“the nightly dream registers in Task Scheduler for your own user (no admin rights)”— readme
“a nightly "dreaming" process that learns from your conversations”— readme
“nightly dreaming all work — the nightly dream registers in Task Scheduler”— readme
“Give your AI a memory, habits, and a daily routine — with nothing but files.”— readme
“memory and self-improving skills for Claude Code/Codex”— topics
信号来源: Show HN
访问官网 →
📛 官方徽章挂到官网 / README
你是 Wienerdog 的作者?上面选样式,嵌入代码实时更新;要贴合站点配色可在 URL 上加 bg= / fg= / accent=(十六进制色)。徽章链回本页。
手机端点「分享」直达微信/朋友圈/小红书;桌面端用「复制文案」后到 App 内粘贴发布
常见问题
Wienerdog 是什么?
为 Claude Code 和 Codex 提供记忆与自我进化能力的 AI 助手，能记住过往交互、持续优化编程建议，提升开发效率
Wienerdog 解决什么问题?
在编程过程中记住用户习惯与历史上下文，持续优化代码建议与自动补全
Wienerdog 为什么值得关注?
突破传统 AI 编程助手无记忆的局限，实现长期学习与上下文延续，是 AI agent 的典型应用
Wienerdog 是开源的吗?
是。Wienerdog 为开源项目。
Wienerdog 能自托管吗?
可以。Wienerdog 支持自托管部署。
Wienerdog 有多少人在用?
KanonAgent 记录到:8 票(本站首次收录于 2026-08-01)。
Wienerdog 有什么替代品?
KanonAgent 库内的同类 agent:Stealth Venture、taste-skill、archify、open-design、deer-flow、ruflo。