第43期 | Our response to the Axios developer tool c
今日摘要
LangChain Blog：Your harness, your memory Harrison's In the Loop Series 7 min read Apr 11, 2026 Agent harnesses are becoming the domina…
Simon Willison：I think it's non-obvious to many people that the OpenAI voice mode runs on a much older, much weaker model - it feels l…
OpenAI Blog：April 10, 2026 Security Our response to the Axios developer tool compromise Loading… Share We recently identified a sec…
Simon Willison：Check the size of any GitHub repository by entering the owner and repository name or pasting a GitHub URL. The tool fet…
Simon Willison：gzip middleware for ASGI applications, extracted from Starlette 9th April 2026 I ran into trouble deploying a new featu…
总结 + 观点：Your harness, your memory Harrison's In the Loo…｜中文观点：Your harness, your memory 的价值在于它是否能真正降低智能体落地门槛，而不是…
总结 + 观点：I think it's non-obvious to many people that th…｜中文观点：围绕 ChatGPT voice mode is a weaker model，真正重要的是它会不会…
总结 + 观点：April 10, 2026 Security Our response to the Axi…｜中文观点：从 Our response to the Axios developer tool comprom…
总结 + 观点：Check the size of any GitHub repository by ente…｜中文观点：GitHub Repo Size 的核心不在新鲜感，而在它是否能提升工程效率、部署稳定性或开发者工作…
总结 + 观点：gzip middleware for ASGI applications, extracte…｜中文观点：对 asgi-gzip 0.3 来说，更值得判断的是它会不会进入团队默认工具链，而不是短期讨论热度。
Your harness, your memory
标签：#ai_engineering_blogs #core
作者：
原文：Your harness, your memory Harrison's In the Loop Series 7 min read Apr 11, 2026 Agent harnesses are becoming the dominant way to build agents, and they are not going anywhere. These harnesses are intimately tied to agent memory.
ChatGPT voice mode is a weaker model
标签：#ai_engineering_blogs #core
作者：
原文：I think it's non-obvious to many people that the OpenAI voice mode runs on a much older, much weaker model - it feels like the AI that you can talk … 10th April 2026 I think it's non-obvious to many people that the OpenAI voice mode runs on a much older, much weaker model - it feels like the AI that you can talk to sho...
链接：https://simonwillison.net/2026/Apr/10/voice-mode-is-weaker/#atom-everything
Our response to the Axios developer tool compromise
标签：#ai_engineering_blogs #core
作者：
原文：April 10, 2026 Security Our response to the Axios developer tool compromise Loading… Share We recently identified a security issue involving a third-party developer tool, Axios, that was part of a widely reported, broader industry incident (opens in a new window) . Out of an abundance of caution we are taking steps t...
GitHub Repo Size
标签：#ai_engineering_blogs #core
作者：
原文：Check the size of any GitHub repository by entering the owner and repository name or pasting a GitHub URL. The tool fetches repository data from the GitHub API and displays … 9th April 2026 GitHub doesn't tell you the repo size in the UI, but it's available in the CORS-friendly API .
链接：https://simonwillison.net/2026/Apr/9/github-repo-size/#atom-everything
asgi-gzip 0.3
标签：#ai_engineering_blogs #core
作者：
原文：gzip middleware for ASGI applications, extracted from Starlette 9th April 2026 I ran into trouble deploying a new feature using SSE to a production Datasette instance, and it turned out that instance was using datasette-gzip which uses asgi-gzip which was incorrectly compressing event/text-stream responses. asgi-gzip w...
链接：https://simonwillison.net/2026/Apr/9/asgi-gzip/#atom-everything
Previewing Interrupt 2026: Agents at Enterprise Scale
标签：#ai_engineering_blogs #core
作者：
原文：This year, we're doing it again. Interrupt 2026 is May 13–14 at The Midway in San Francisco, and the lineup, the format, and the scale have all leveled up.
链接：https://blog.langchain.com/previewing-interrupt-2026-agents-at-enterprise-scale/
Deep Agents Deploy: an open alternative to Claude Managed Agents
标签：#ai_engineering_blogs #core
作者：
原文：Today we’re launching Deep Agents deploy in beta. Deep Agents deploy is the fastest way to deploy a model agnostic, open source agent harness in a production ready way.
链接：https://blog.langchain.com/deep-agents-deploy-an-open-alternative-to-claude-managed-agents/
Human judgment in the agent improvement loop
标签：#ai_engineering_blogs #core
作者：
原文：AI agents work best when they reflect the knowledge and judgment your team has built over time. Some of that is institutional knowledge that’s already documented and easy for an agent to use as-is.
链接：https://blog.langchain.com/human-judgment-in-the-agent-improvement-loop/
Multimodal Embedding & Reranker Models with Sentence Transformers
标签：#ai_engineering_blogs #core
作者：
原文：Multimodal Embedding & Reranker Models with Sentence Transformers
链接：https://huggingface.co/blog/multimodal-sentence-transformers
OpenAI Full Fan Mode Contest: Terms & Conditions
标签：#ai_engineering_blogs #core
作者：
原文：Explore the official terms and conditions for the OpenAI Full Fan Mode Contest, including eligibility, entry steps, judging criteria, and prize details. Learn how to participate, submit your entry on Instagram, and win IPL match tickets.
链接：https://openai.com/index/full-fan-mode-contest-terms-conditions