ttys22

#日本語勉強 #Golang #Rust #Archlinux #Fedora

[Tibo @thsottiaux]

More opensource goodness. We have just released a CLI and TypeScript SDK for finding, validating, and fixing security vulnerabilities in your code. Scan repositories, review changes, track findings over time, and run security checks in CI.

https://t.co/nkfTbw8p7b

[宝玉 @dotey]

MCP 协议今天推出了第五个大版本（版本号 2026-07-28），核心变化：从有状态双向协议变成了无状态的请求/响应协议。

这是 MCP 社区呼声最高的改动。

之前的 MCP 客户端连上来要先握手，服务端要给你一个会话 ID（session ID），后续每次请求都得带着这个 ID。这意味着你的请求被���制在了某一台服务器实例上，不方便做负载均衡。

现在每个请求都是独立的，自带协议版本和客户端信息，可以负载到任何一个实例上。也就是说 MCP 服务器终于可以像普通 HTTP 服务一样部署了，serverless、边缘计算、或者 CDN 后面放一排实例。

那如果业务确实需要跨请求的状态呢？协议的建议是：由工具自己生成一个句柄（handle），让模型在工具调用之间传递。状态放在业务层而不是协议层，模型能看到这个句柄，也能理解要怎么用它。

【其他变化】

MRTR（多轮往返请求）解决中途需要用户输入确认的问题。

比如 Supabase 的 MCP 服务器想在创建项目前告诉你费用，或者在执行删除操作前让你确认。以前这需要维持一个双向流不断开，现在服务端返回“需要输入”状态，客户端带上用户的回答重新发请求即可。

请求头里新增了 Mcp-Method 和 Mcp-Name 两个字段，网关和防火墙可以直接根据请求头做路由和鉴权，不用解析 JSON 请求体。

授权方面做了几项加固，包括要求验证授权服务器的 issuer 参数（堵上了一个授权服务器混淆漏洞）。动态客户端注册（DCR）正式废弃，改用客户端元数据文档（CIMD）。

Roots、Sampling、Logging 三个功能标记为废弃，但至少还能用 12 个月。旧版 HTTP+SSE 传输同样进入废弃轨���。协议首次引入正式的废弃策略，保证最少 12 个月过渡窗口，这对生产环境的团队来说很实际：你可以排期升级，不用被动救火。

【生态与迁移】

四个一级开发工具包（SDK）同步更新：TypeScript、Python、Go、C#。Rust SDK 以 beta 状态跟进。MCP 目前月下载量接近 5 亿次，TypeScript 和 Python SDK 各自累计下载量都超过了 10 亿。

AWS（Amazon Bedrock AgentCore）、Microsoft（Foundry）、Cloudflare（Workers）、Google Cloud 等都已宣布支持新规范。Figma、Supabase、Honeycomb 等工具厂商也在跟进。

新规范有破坏性变更，依赖会话标识符的实现需要改造。SDK 团队表示根据早期测试反馈已经简化了迁移流程，但具体成本因项目而异。如果你现在在生产环境跑着 MCP 服务器，建议先看一遍完整的 changelog 和迁移指南，评估影响面再动手。

[GitHub @github]

You're not behind. There's no secret everyone else has.

There's just the harness, and it's mostly all you need.

@burkeholland gives you a simple, repeatable workflow for GitHub Copilot. https://t.co/1gvgpf0ioi

[ClaudeDevs @ClaudeDevs]

MCP 2026-07-28 is live and it's the largest update to the protocol since launch.

MCP is now stateless, making it easier to deploy and scale remote servers.

https://t.co/K8KqxbUh4e

[Elon Musk @elonmusk]

Interesting.

Grok 4.6 releases around August 7. This will be the 1.5T model with significantly improved SFT & RL.

Grok 4.7 will be the 2.1T model released a few weeks later. This will be better than 4.6 in every way, except slightly slower to serve, albeit with even better token efficiency.

[Kimi.ai @Kimi_Moonshot]

We've open-sourced AgentENV in collaboration with kvcache-ai.

AgentENV is a distributed system for running agent environments at scale. Its components power agentic RL training for Kimi K3, with fast snapshot, resume, and fork support for large-scale parallel agent workflows.

Explore on GitHub:
https://t.co/Dsxyw5rLqm

[ttys22 @ttys22]

Oops, thanks @cursor_ai confirmed #Opus 5 for me

[Google DeepMind @GoogleDeepMind]

We’re rolling out three new models to make AI agents faster, smarter, and cheaper at scale:

🔵 Gemini 3.6 Flash: It uses fewer tokens than 3.5 Flash to deliver higher quality work at the exact same cost.

🔵 Gemini 3.5 Flash-Lite: A fast, cost-effective option for everyday tasks like processing documents and agentic search.

🔵 Gemini 3.5 Flash Cyber: A cybersecurity model built to find and patch critical software vulnerabilities.

[Logan Kilpatrick @OfficialLoganK]

We have started our most ambitious pre-training run yet, for Gemini 4, and are excited by the progress : )

[Logan Kilpatrick @OfficialLoganK]

@estebs Gemini 3.5 Pro is currently testing with partners and we plan to make it broadly available as soon as it’s ready

[ttys22 @ttys22]

#Gemini 3.6 flash

https://t.co/eExUBGAC5z

[Google Antigravity @antigravity]

Learn more: https://t.co/aC7tiyLka6

[Google Antigravity @antigravity]

Gemini 3.6 Flash is live in Antigravity! ⚡️

Building on 3.5 Flash feedback, it consumes up to 17% fewer output tokens while completing complex workflows in fewer reasoning steps and tool calls.

[Entelligence AI @EntelligenceAI]

Best AI models by use case (July 2026):

🎨 Frontend → Kimi-K3
⚙️ Backend → Claude Fable 5
🐛 Complex Debugging → GPT-5.6 Sol
🖼️ Image Generation → GPT
🌍 Translation → Gemini 3.5 Flash
🔎 Real-time Search → Grok 4.5
🎥 Video Generation → Seedance 2.0
💰 Best Value → DeepSeek V4 Pro
🖥️ Local AI (Large) → GLM-5.2
💻 Local AI (Small, ~128GB RAM) → HY-3 / DeepSeek V4 Flash

One thing I've learned over the past few months:

Stop looking for one model that does everything.

The biggest productivity gains come from routing each task to the model that's best at it.

The AI stack is becoming like a software stack.

Different tools.
Different strengths.
Much better results.

Bookmark this.

[宝玉 @dotey]

靴子落地了：从7月20日起，Claude Fable 5 将包含在所有 Max 和 Team Premium 计划中，限额为50%。

这得感谢 GPT 5.6 和 Kimi 3

[ttys22 @ttys22]

@cyandev 并不是, 现在也只有 A 社还没改进. Gemini 已经有了
Interactions API https://t.co/elpzWD1ebS

https://t.co/CGVPDoSqdm

[Arena.ai @arena]

Kimi-K3 just topped the Frontend Code Arena with a 76% pairwise win rate.

When its output was compared head-to-head against other models on the same task, it was picked as the better output 76% of the time on average.

For reference: Claude Fable 5 (63%), GPT-5.6 Sol (58%). 50% is baseline, a model winning and losing equally often.

[SpaceXAI @SpaceXAI]

We've open-sourced Grok Build and have reset usage limits for all users.

Open sourcing Grok Build allows anyone to support making a reliable and robust harness. Check out our code, including the Git repo for the Grok Build CLI.

https://t.co/3SSvPu2Nrz

[Google AI Developers @googleaidevs]

The Interactions API is now GA. Featuring a stable schema and new capabilities via a single /interactions endpoint, it is now our standard interface for both model inference and autonomous agents.

Check out the key updates:

— Managed Agents: Spin up a secure, remote Linux sandbox for code execution and browsing via a single API call, using the default antigravity-preview-05-2026 agent or your own custom instructions

— Simplified Schema: Every action is now represented as its own step (from user_input to model_output, etc.) in a simplified, linear flow

— Background Execution: Set background=True to run models and agents asynchronously, letting you poll for results when ready

— Expanded Tool Ecosystem: Native support for Computer Use, File Search, and Google Maps, alongside new agent-building tools like the Gemini API Docs MCP server and gemini-interactions-api Skill

— Media Generation: Image generation via Google Search-grounded Nano Banana 2, music with Lyria 3, and expressive multi-speaker TTS

Read the blog to learn more: https://t.co/ZzfpUGzAvm

[Jarred Sumner @jarredsumner]

Rewriting Bun in Rust

https://t.co/Rl8bcaxBFc