wanshuiyin

Activity

Stars and mentions · Last 30 days, Oct 9 – Nov 7

GitHub stars1,165

Mentions4

UTC days · repositories wanshuiyin owns6 repositories

Top repositories

wanshuiyin’s repositories ranked by new forks, as on trending.

#	Repository	Activity ⓘ	Stars	Forks	Contributors
1
2
3
4
5
6
7
8
9
10

Contributes to

Repositories outside wanshuiyin’s account where wanshuiyin opened or reviewed pull requests in the past 365 days.

Who noticed wanshuiyin

Everyone who starred or forked wanshuiyin’s repositories since the archive began in 2011.

Notable accounts

Ordered by the stars their own repositories have received.

Biggest fans

Accounts that starred the most of these repositories.

Breakout weeks

Each repository’s best week of new stars, with the release published around it.

Mentions

5posts

Nov 81 post

X · Nov 8 · @WShuiyin · wanshuiyin/SuperARIS
20k 🌟ARIS 再次开源SuperARIS：一行搭起主动多Agent 协同Personal 科研平台之前开发的ARIS(Auto Resrach In Sleep)虽然能够通过mcp的形式(我也推荐多智能体协作，质量明显高)但是不直观所以基于herdr开发了SuperARIS🥳🥳🥳 wanshuiyin/SuperARIS 🔥安装好后，在终端输入 superaris，选好项目、团队和模型，就能打开你的科研工作台🔥 Key Feature： 1️⃣主动跟进，让交付接上下一步 2️⃣长程协作，让记忆留在项目里 3️⃣保留 ARIS 的完整流程与 MCP 协作设计SuperARIS时，我希望 Agent能主动推进流程而不是卡住：记得项目做到哪里、谁负责什么、结果回来后该接着做什么。在 SuperARIS 里，你把目标交给主 Agent，它负责拆解任务、组织分工、回收和核验成果。正如我之前所说，让 Codex 常驻做工程：实现基线、跑实验、核对结果；让 Gemini 参与想法讨论和论文主写。每个成员都有自己的会话，你也可以随时进去直接共创、修改或接手。还有一些我觉得比较重要的feature 1️⃣主动跟进，让交付接上下一步项目运行期间，主 Agent 能接收任务进展和协作者交付，核对成果，再安排后续工作。实验有了结果，就可以交给写作成员组织论证；审阅发现问题，再回到工程或正文修改。你把握研究方向，团队持续推进具体工作。 2️⃣长程协作，让记忆留在项目里常驻成员保留各自的讨论上下文，围绕同一个课题持续工作。目标、决定、纠正、实验结果和下一步，保存在项目笔记与研究知识库中。跨天、跨会话回来时，可以接着已有进展继续做。 3️⃣保留 ARIS 的完整流程与 MCP 协作 ARIS 原有的workflow、skills 和 MCP 能力继续保留。你可以推进完整科研流程，也可以把一段 workflow 分给某个成员，或者只调用一个 skill。每个成员都能按任务读取 ARIS 技能与项目上下文。常驻团队负责持续协作；原有的 MCP 临时 worker 和独立 reviewer 继续按需调用。需要额外执行能力或全新上下文的审阅时，依然保留只有 Codex 也能用，也支持 cc + Codex，或者额外agy也行
9 likes · 4,994 views
X · Nov 3 · @Xudong07452910 · wanshuiyin/Auto-claude-code-research-in-sleep
@yhlyjs @Macro_Zyaire 这个也还不错wanshuiyin/Auto-claude-code-research-in-sleep
2 likes · 56 views
X · Nov 1 · @WShuiyin · wanshuiyin/HERO-Anti-OverDefense
@realTaki42 @bot @ChatGPT 防御性编程和长程goal的遗忘简直是codex 通病hhh，可以看看HERO wanshuiyin/HERO-Anti-OverDefense 针对这两个问题都有解决🤓🤓
1 like · 103 views
X · Oct 22 · @OsamaGweaa · wanshuiyin/Auto-claude-code-research-in-sleep
@LeoBuilds_ This repo connects both Claude & codex Claude’ strength is fast, fluid execution; Codex is slower but more deliberate and rigorous in critique. These complementary styles speed × rigor produce better outcomes than either model talking to itself. wanshuiyin/Auto-claude-code-research-in-sleep
22 views
X · Oct 11 · @WShuiyin · wanshuiyin/Auto-claude-code-research-in-sleep
17k🌟自主科研ARIS|开源国模(Deepseek V4pro-preview + GLM 5.1)协作超越5.6Sol 转眼间ARIS(Auto Research in Sleep，wanshuiyin/Auto-claude-code-research-in-sleep)已经14000 star了🎉 最近可算抽出来时间把ARIS在各类AutoResearch 的Benchmark上面测了测😋😋 结果还是挺出乎我意料的，发现在已有的AutoResearch Bencmark上，ARIS使用DeepSeekV4 Pro和GLM5.1(这也是ARIS的平替建议组合)根据workflow进行对抗性配合和协作就能够轻松追平GPT5.6 sol🔥🔥 这说明对于比较长程Auto Research的任务来说，是不是任务执行结果的可验证和可回溯比纯粹的模型能力更重要？只要配合的好，未必不能做到更好🧐🧐 (btw国模真的太强了，我都不敢想最先进的国模能达到什么境地，有时间测测国内各大玩家的模型配合的怎么样，感觉可以给我很多惊喜啊🤩🤩 希望可以和大家多交流！ ARIS也在持续更新，我们最近也更新了针对Theory Proof的workflow 和lean 验证以及自己独立的ARIS-Code，欢迎大家给我提issue和讨论🙇🏻‍♂️🙇🏻‍♂️
1 like · 1,020 views

Achievements

Awards earned by the repositories wanshuiyin owns. Select an award to embed its badge.

No awards yet

Repositories earn awards by reaching the top three in a completed trending day or week.

A repository earns an award for finishing first, second or third in a completed trending day or week, ranked by new stars, new forks or pull request authors. Awards follow a repository through renames and transfers and count toward its current owner. Weeks run Monday to Sunday, UTC.