How-toMulti-source7月27日 12:13

bento PPT Skill：输入主题自动生成HTML PPT

向阳乔木@vista8

Post· 11 sources

View the post

Why we picked it

用npx安装这个Skill，输入主题就能生成可编辑的HTML PPT，还能在线协作，推荐搭配Kimi K3或Opus 4.8+使用。

该Skill基于bento PPT开发，输入内容或主题即可自动生成可编辑、可在线演示、可协作的HTML PPT。

Body

安装指令为npx skills add joeseesun/qiaomu-bento-ppt。开发者推荐使用Kimi K3或Opus 4.

8+等前端审美较好的模型来获得更佳效果。该项目已在GitHub开源。

Original · 向阳乔木

基于 bento PPT 改造了一个 Skill。输入内容或主题，自动生成可编辑修改、在线演示，甚至可以发给别人协作的HTML PPT。安装指令：npx skills add joeseesun/qiaomu-bento-ppt 推荐用前端审美好的模型，比如Kimi K3或Opus 4.8+ 开源见评论区📊 6 ⚡

Industry Coordinates & Source Verification

This report is categorized under AITOP's "How-to" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include bento PPT, HTML PPT, Skill, Kimi K3.

Source attribution: curated from 向阳乔木 (@vista8). Cross-source corroboration: 11 independent sources have covered this development (including elvis, Aadit Sheth, AI Will), confirming industry-wide signal strength.

elvis07-27 15:42Source
Aadit Sheth07-27 18:16Source
AI Will07-26 10:00Source
歸藏(guizang.ai)07-27 08:58Source
IT之家07-27 15:14Source
Yangyi07-27 15:26Source
AK07-27 15:29Source
Fireworks AI07-27 15:30Source
Cognition07-27 15:38Source
Guillermo Rauch07-27 15:38Source

#bento PPT #HTML PPT #Skill #Kimi K3 #Opus 4.8+

Keep reading

Recent stories with the same tag or section

How-to Builders01:24

Chinese contentPlaud Note Pro 体验：手机有录音和 AI，为什么还需要一台录音设备

如果你天天开会、采访，手机录音老是被来电打断，这台 Plaud Note Pro 能挂身被动录，再用 AI 自动出摘要。它不像手机那样分心，适合想偷懒记笔记的人。

#Plaud Note Pro #meeting notes #productivity tools

GenkSource

How-to Builders01:06

Chinese contentStanford 教授 Marco Pavone 自主机器人教材与课程资料免费公开

Stanford 教授 Marco Pavone 把自主机器人的整本教材加课程资料全放网上了，一分钱不收。想入行机器人或自动驾驶的人不用再到处拼凑资料，跟着 Stanford 正课的讲义走一遍就行，比大多数零散视频教程系统得多。

#Stanford #autonomous driving

Stanford HAI@StanfordHAISource

How-to BuildersMulti-source01:05

Chinese contentZed 让智能体按需查阅生成代码的原始对话线程

Zed 这个细节挺值得琢磨：代码是哪段对话写的，智能体能查摘要、再按需翻原始讨论，但不自动灌满上下文。比起 Cursor 一把梭全塞进窗口，这种按需取用的方式对长会话开发者更有参考价值。

#Zed #coding assistants #context management

Event

Zed@zeddotdevSource

How-to Builders01:03

Chinese contentGergely Orosz：AI 服务不稳定，单一订阅行不通

Pragmatic Engineer 的作者 Gergely 分享了他的真实用法：Claude 挂了就切到 Grok Bot 干活。如果你每天重度依赖某一家 AI 订阅，这种多订阅轮换思路能帮你少踩坑。

#Claude #Grok #AI tools

Gergely Orosz@GergelyOroszSource

How-to BuildersMulti-source01:02

Chinese contentsmolmaxxing：让1G内存的旧电视盒子和iPad也能本地跑大模型

家里吃灰的电视盒子、旧iPad现在真能跑本地大模型了，1G内存都行，实测token速度都贴出来了。用的还是Qwen3.5、Ling-mini-2.0这些模型，源码在GitHub上，想白嫖旧硬件玩本地推理的可以照着改。

#smolmaxxing #llama.cpp #Qwen3.5 #local deployment

Event

Gorden Sun@Gorden_SunSource

How-to BuildersSelected01:00

SparseRead Cuts Agent Reading Tokens by Up to 92.9%

The shift from post-hoc context compression to pre-read filtering changes how agent harnesses should be designed, and the reported 92.9% token saving suggests the approach scales beyond toy demos. Builders working on coding agents, RAG pipelines, or long-horizon automation can adapt the intent-locate-budget loop directly into their own tooling. Compared with simply adding search tools or summarisation layers, SparseRead treats reading itself as a budgeted, task-driven control problem.

#SparseRead #Claude Code #agents #Agent Harness

AIGCLINK@aigclinkSource