X好文分享，技术类
- 内容介绍
- 文章标签
- 相关推荐
业务agent化 https://x.com/xingpt/status/2025219080421277813?s=46
自动读推文 https://x.com/0xkevinhe/status/2025783766741123193?s=46
学东西快十倍的方法 https://x.com/phosphenq/status/2026383675735048248?s=46
skill全览 https://x.com/GoSailGlobal/status/2030490279221051487
自动科研雏形 https://x.com/karpathy/status/2030371219518931079
OC上下文管理器 https://x.com/xxx111god/status/2030661557873451370
claude skill 2.0 https://x.com/itsolelehmann/status/2031679839476875734?s=46
nano banana2官方演示 https://x.com/GeminiApp/status/2031784342037680413
Google AI设计助手 https://x.com/stitchbygoogle/status/2034332847893574080
Telegram接入claude code 官方通道 https://x.com/trq212/status/2034761016320696565
谷歌内存论文破解 https://x.com/k1rallik/status/2038567965465944491?s=46
--【壹】--：
感谢佬友
--【贰】--： ricerug:
OC上下文管理器
Jason Zuo
@xxx111god 有人问这个可视化怎么做的 说实话就是一个简单的openclaw context detail的wrapper script 把打包成了一个🦞OpenClaw skill：context-doctor 可以直接用 clawhub install context-doctor 一行装完，自动扫 workspace 文件状态、token 占用、bootstrap 占比 三个诊断信号： 🔴 TRUNCATED — pic.x.com/Kka4L3KlWH x.com/xxx111god/stat…Jason Zuo @xxx111god
之前聊了怎么把 bootstrap context 砍 75% 但砍完之后怎么验证？怎么知道你的 agent 有没有在"盲飞"？ OpenClaw 有个命令 /context list，能看到每个注入文件的状态和 token 占用。我做了个可视化版本（见图） 三个诊断信号： 🔴 文件状态 TRUNCATED — 指令被截断，agent 在盲飞 🟡 文件状态 https://t.co/zwQ6lDBziq https://t.co/5x1gtUlIQE 3:06 PM - 8 Mar 2026 200 29--【叁】--：
@yukicantcode @NingNingUP
--【肆】--：
爱学习的在哪儿都能学习！
--【伍】--：
这软件还能看正经东西？
--【陆】--： ricerug:
skill全览
这个是真不错
Agent Skills Hub - Discover Agent Skills, Tools & MCP Servers
Discover, compare and explore 9,000+ open-source Agent Skills, AI tools, MCP servers and Claude skills.
--【柒】--： ricerug:
自动科研雏形
这个对于搭建系统非常有启发
Andrej Karpathy
@karpathy I packaged up the "autoresearch" project into a new self-contained minimal repo if people would like to play over the weekend. It's basically nanochat LLM training core stripped down to a single-GPU, one file version of ~630 lines of code, then: - the human iterates on the pic.x.com/3tyOq2P9c6 7:53 PM - 7 Mar 2026 22K 2.7K--【捌】--：
自动研究最新进展
Andrej Karpathy
@karpathy Three days ago I left autoresearch tuning nanochat for ~2 days on depth=12 model. It found ~20 changes that improved the validation loss. I tested these changes yesterday and all of them were additive and transferred to larger (depth=24) models. Stacking up all of these changes, pic.x.com/j34dSt4oht 10:28 PM - 9 Mar 2026 11K 1.1K--【玖】--：
之前有个主题讨论得热火朝天，现在找不到了
--【拾】--：
不错，学习了
--【拾壹】--：
Visual Explainer
Nico Bailon
@nicopreme POV: Planning with the "Visual Explainer" skill. I can't go back to markdown plans after getting used to this. github.com/nicobailon/vis… pic.x.com/m2zz9ynDEn x.com/nicopreme/stat…Nico Bailon @nicopreme
Created an agent skill called “Visual Explainer” + set of complementary slash commands aimed to reduce my cognitive debt so the agent can explain complex things as rich HTML pages. The skill includes reference templates and a CSS pattern library so output stays consistently https://t.co/r1PJ6LLlj2 11:40 PM - 19 Feb 2026 2.7K 170Prompt Caching
Thariq
@trq212 x.com/i/article/2024… 7:57 PM - 19 Feb 2026 5K 470--【拾贰】--：
可怜我的X被封了
--【拾叁】--：
码住了，感谢佬友
--【拾肆】--： ricerug:
banana2演示
Marcio Lima 利真 マルシオ 💎
@Preda2005 @GeminiApp 🗼🗻🌸 pic.x.com/KaE9mkE2CA 9:30 PM - 11 Mar 2026 180 6--【拾伍】--：
收藏了。
--【拾陆】--：
腾讯那个是不是仿照这个做的
--【拾柒】--：
有分享X上的好片的么。。
--【拾捌】--： ricerug:
claude skill 2.0
Ole Lehmann
@itsolelehmann claude skills 2.0 quietly launched this week and most people missed it anthropic shipped 3 upgrades to skills that fix the problems almost everyone runs into here's the breakdown (and how to upgrade yours): problem 1: you had no way to measure how well your skills were x.com/itsolelehmann/…Ole Lehmann @itsolelehmann
https://t.co/jzOGxmkyhx 10:33 AM - 11 Mar 2026 270 24业务agent化 https://x.com/xingpt/status/2025219080421277813?s=46
自动读推文 https://x.com/0xkevinhe/status/2025783766741123193?s=46
学东西快十倍的方法 https://x.com/phosphenq/status/2026383675735048248?s=46
skill全览 https://x.com/GoSailGlobal/status/2030490279221051487
自动科研雏形 https://x.com/karpathy/status/2030371219518931079
OC上下文管理器 https://x.com/xxx111god/status/2030661557873451370
claude skill 2.0 https://x.com/itsolelehmann/status/2031679839476875734?s=46
nano banana2官方演示 https://x.com/GeminiApp/status/2031784342037680413
Google AI设计助手 https://x.com/stitchbygoogle/status/2034332847893574080
Telegram接入claude code 官方通道 https://x.com/trq212/status/2034761016320696565
谷歌内存论文破解 https://x.com/k1rallik/status/2038567965465944491?s=46
--【壹】--：
感谢佬友
--【贰】--： ricerug:
OC上下文管理器
Jason Zuo
@xxx111god 有人问这个可视化怎么做的 说实话就是一个简单的openclaw context detail的wrapper script 把打包成了一个🦞OpenClaw skill：context-doctor 可以直接用 clawhub install context-doctor 一行装完，自动扫 workspace 文件状态、token 占用、bootstrap 占比 三个诊断信号： 🔴 TRUNCATED — pic.x.com/Kka4L3KlWH x.com/xxx111god/stat…Jason Zuo @xxx111god
之前聊了怎么把 bootstrap context 砍 75% 但砍完之后怎么验证？怎么知道你的 agent 有没有在"盲飞"？ OpenClaw 有个命令 /context list，能看到每个注入文件的状态和 token 占用。我做了个可视化版本（见图） 三个诊断信号： 🔴 文件状态 TRUNCATED — 指令被截断，agent 在盲飞 🟡 文件状态 https://t.co/zwQ6lDBziq https://t.co/5x1gtUlIQE 3:06 PM - 8 Mar 2026 200 29--【叁】--：
@yukicantcode @NingNingUP
--【肆】--：
爱学习的在哪儿都能学习！
--【伍】--：
这软件还能看正经东西？
--【陆】--： ricerug:
skill全览
这个是真不错
Agent Skills Hub - Discover Agent Skills, Tools & MCP Servers
Discover, compare and explore 9,000+ open-source Agent Skills, AI tools, MCP servers and Claude skills.
--【柒】--： ricerug:
自动科研雏形
这个对于搭建系统非常有启发
Andrej Karpathy
@karpathy I packaged up the "autoresearch" project into a new self-contained minimal repo if people would like to play over the weekend. It's basically nanochat LLM training core stripped down to a single-GPU, one file version of ~630 lines of code, then: - the human iterates on the pic.x.com/3tyOq2P9c6 7:53 PM - 7 Mar 2026 22K 2.7K--【捌】--：
自动研究最新进展
Andrej Karpathy
@karpathy Three days ago I left autoresearch tuning nanochat for ~2 days on depth=12 model. It found ~20 changes that improved the validation loss. I tested these changes yesterday and all of them were additive and transferred to larger (depth=24) models. Stacking up all of these changes, pic.x.com/j34dSt4oht 10:28 PM - 9 Mar 2026 11K 1.1K--【玖】--：
之前有个主题讨论得热火朝天，现在找不到了
--【拾】--：
不错，学习了
--【拾壹】--：
Visual Explainer
Nico Bailon
@nicopreme POV: Planning with the "Visual Explainer" skill. I can't go back to markdown plans after getting used to this. github.com/nicobailon/vis… pic.x.com/m2zz9ynDEn x.com/nicopreme/stat…Nico Bailon @nicopreme
Created an agent skill called “Visual Explainer” + set of complementary slash commands aimed to reduce my cognitive debt so the agent can explain complex things as rich HTML pages. The skill includes reference templates and a CSS pattern library so output stays consistently https://t.co/r1PJ6LLlj2 11:40 PM - 19 Feb 2026 2.7K 170Prompt Caching
Thariq
@trq212 x.com/i/article/2024… 7:57 PM - 19 Feb 2026 5K 470--【拾贰】--：
可怜我的X被封了
--【拾叁】--：
码住了，感谢佬友
--【拾肆】--： ricerug:
banana2演示
Marcio Lima 利真 マルシオ 💎
@Preda2005 @GeminiApp 🗼🗻🌸 pic.x.com/KaE9mkE2CA 9:30 PM - 11 Mar 2026 180 6--【拾伍】--：
收藏了。
--【拾陆】--：
腾讯那个是不是仿照这个做的
--【拾柒】--：
有分享X上的好片的么。。
--【拾捌】--： ricerug:
claude skill 2.0
Ole Lehmann
@itsolelehmann claude skills 2.0 quietly launched this week and most people missed it anthropic shipped 3 upgrades to skills that fix the problems almost everyone runs into here's the breakdown (and how to upgrade yours): problem 1: you had no way to measure how well your skills were x.com/itsolelehmann/…Ole Lehmann @itsolelehmann
https://t.co/jzOGxmkyhx 10:33 AM - 11 Mar 2026 270 24