1
Adobe Photoshop 推出“AI 辅助编辑器”测试版：基于提示词的全新图像编辑模式
IT之家 8 月 27 日消息，Adobe 今日宣布为 Photoshop 新增“AI Assisted Editor”（AI 辅助编辑器）测试体验。 该功能采用基于自然语言提示词的编辑方式，提供简化版操作界面，用户无需使用传统复杂工具即可完成图像修改。 AI 辅助编辑器作为 Photoshop 中…
2
Show HN: Crilio – pytest for prompts
用pytest风格测试LLM提示词，Crilio让AI输出验证变得可编程。
Article URL: https://github.com/mukundzha/crilio Comments URL: https://news.ycombinator.com/item?id=49433864 Points: 5 # Comments: 0
3
PromptResponse: Optimizing Prompts for LLM Coding Tasks
格式与提示词微观调整如何影响大模型编码表现？这项对照实验给出可复现的优化路径。
arXiv:2608.21074v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used in research workflows and software development pi…
4
Beyond Prompt Engineering: A Systematic Analysis of Prompt Lexical Sensitivity and Its Impacts on Quality
同一提示词换个词性能骤变，这项研究首次从n-gram词元级别系统拆解LLM的提示词敏感性，帮你避开微调陷阱。
arXiv:2608.20349v1 Announce Type: cross Abstract: Large Language Models (LLMs) exhibit extreme sensitivity to surface-level prompt variations, in whic…
5
SCOPE: A Generative Approach for LLM Prompt Compression
LLM提示词压缩新思路：生成式方法SCOPE，有效压缩上下文同时保留性能，COLM 2026收录。
arXiv:2508.15813v2 Announce Type: replace-cross Abstract: A big issue in modern LLM applications is they tend to feed long context to LLM, which resul…
6
Stop chasing fads. Operate Claude like a startup founder.
用创业者的成本思维使用Claude：缓存不变的系统提示词与工具定义，输入成本能降到约十分之一。
I am running an AI-native startup. Early on though, I would spend so much time staying ahead of every new AI hot take that I lost focus on what I need…
7
Claudette: Make Claude stop talking like a BuzzFeed article
受够了AI的标题党腔调？这个开源项目让Claude回归正经说话。
Article URL: https://github.com/adnanakil/nobuzz/blob/main/README.md Comments URL: https://news.ycombinator.com/item?id=49388752 Points: 251 # Comment…
8
Show HN: Muteval – mutation testing for your LLM evals
变异测试一招揪出LLM评测漏洞，Muteval让评估数据更可靠，开源免费直接上手。
Article URL: https://github.com/AshwinUgale/muteval Comments URL: https://news.ycombinator.com/item?id=49388918 Points: 1 # Comments: 0
9
CoSnitch Is a Reminder That Your Chatbot Will Tell on You If You Ask Nicely Enough
别再把AI泄密当玩笑，聊天机器人能被套出系统提示词，这不是越狱而是侦察。
An AI assistant got talked into describing its own guts to a stranger. That's not a jailbreak curiosity, that's reconnaissance-as-a-service, and it sh…
10
Show HN: I made a comic series explaining Claude – Chat, Cowork, and Code
用漫画轻松搞懂Claude的聊天、协作与编程，零基础也能秒懂AI用法
Article URL: https://github.com/Fake4d/claude-school Comments URL: https://news.ycombinator.com/item?id=49356989 Points: 1 # Comments: 0
11
Whether LLMs Can Navigate Beliefs and Facts Depends on How You Phrase It
同一问题换个问法，大模型对事实与信念的判断就不同，这份研究揭示了提示词如何影响模型的认知边界。
arXiv:2608.17809v1 Announce Type: new Abstract: Humans naturally form and express beliefs in daily communication, e.g., "I think the answer is 3" or "…
12
13
The Value of a Prompt: An LLM-Relative Kolmogorov-Complexity Approach
用列文-柯尔莫哥洛夫复杂度度量提示词价值，给“好提示”一个可计算的理论标尺。
arXiv:2608.16438v1 Announce Type: new Abstract: In a world where valuable artifacts are increasingly created, completed, or processed by LLMs, the cen…
14
CAPO: Constraint-Aware Prompt Optimization for LLM Agents
针对LLM智能体的提示词优化新方法，把约束条件纳入优化过程，让Agent更懂规则、少犯错。
arXiv:2608.16068v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed as agents that rely on system prompts to use …
15
MALDA – a language where LLM prompts and tools are syntax
把大模型提示词写成语言语法，用编程方式直调AI工具，新思路值得一看
Article URL: https://github.com/amaldini/maldalang Comments URL: https://news.ycombinator.com/item?id=49331723 Points: 2 # Comments: 0
16
Don't classify. Hallucinate!
用大模型“幻觉”替代死板分类，自动生成标签，让内容管理更灵活高效。
Don't classify. Hallucinate! I still have quite a bit of older content on my blog that I never got round to tagging. My blog has 1,856 tags - lik…
17
科技爱好者周刊（第 408 期）：你需要知道的 AI 缓存知识
AI缓存让重复对话成本骤降，命中价仅为未命中的五十分之一，十分钟内免重算，省钱关键必看。
这里记录每周值得分享的科技内容，周五发布。...
18
Write your first prompt with the GitHub Copilot app
手把手教你在GitHub Copilot应用中写出第一个提示词，快速上手AI编程助手。
Learn how to write your first prompt in the GitHub Copilot app, choose the right context and model, and start your first task with confidence. The pos…
19
How to Ask the AI: A User Perspective Survey for Large Language Model Prompting
别只顾着问AI，这项用户视角调查告诉你提示词到底该怎么写更有效
arXiv:2608.07494v1 Announce Type: cross Abstract: AI tools like ChatGPT and DeepSeek, powered by Large Language Models (LLMs), allow users to obtain i…
20
I got 30% better AI coding results
实测让 AI 编程效果提升 30% 的代码瘦身法，开源工具 CodeSlimmer 帮你喂给大模型更精准的上下文
Article URL: https://github.com/akasula09/CodeSlimmer Comments URL: https://news.ycombinator.com/item?id=49236555 Points: 2 # Comments: 0