Newsletter 22 August 2026
How this is put together. This page is filled automatically every day from public sources (X, Hacker News, GitHub, YouTube). The headlines come from the source, the Dutch sentence underneath is machine-written. Click through to the source before you base anything on it.
Replit
- This Week in Replit 1) Free Mode: create up to 30X more 2) Conversations: start any idea by chatting 3) Agent modes renamed 4) Routines + Me
- Add Gemini-powered tender search agent (Node/Express + MySQL + frontend)
- Replit Free Mode chat is like an incredibly powerful ChatGPT where the agent has access to a full cloud and it can build anything.
- Well… just like that I hit @Replit level 3 An absolute coincidence that the free agent has dropped and I’ve gone up 3 levels…. Think I need
Claude Code & agents
- Claude Code es un rollo. Hasta que instalas esto. Hay un plugin oficial de Anthropic llamado claude-code-setup. Te dice qué automatizaciones
- If you are using Claude Code without repos. You are missing 50% of what it can do. These are saved skills and plugins the community has buil
- Anthropic Investor Blasts Traditional Venture Capital
- still teaching your coding agent CrewAI patterns by hand? finally. CrewAI now ships official skills for Claude Code, Cursor, Codex, and Wind
Cloudflare OS & self-hosted
- The Only Self-Hosted Apps You Actually Need in 2026
- Our self-hosted Langfuse stack stored 14GB on disk, only 1.6GB of it was actual observation data
- Our self-hosted Langfuse stack stored 14GB on disk, only 1.6GB of it was actual observation data
- I built a companion app for Hermes chat with your agent from your phone, fully self-hosted (open source)
Agent tooling & MCP
- 🚨📊 **SPLUNK PATCHES 17 SECURITY FLAWS — CRITICAL MCP SERVER RCE HITS CVSS 9.1** **CyberSignal Daily ✓ · 🔥 Critical Vulnerability · August 20
- Tencent open-sourced AI-Infra-Guard v4.5.2, a red-teaming platform with 2,000+ CVE rules, .pyc bypass detection, and MCP-Scan RCE prevention
- AI-Infra-Guard - A full-stack AI Red Teaming platform securing AI ecosystems via Agent Scan, Skills Scan, MCP scan, AI Infra scan and LLM ja
- [P2] CVE-2026-32871 on [email protected]
Models & pricing
- OpenAI cuts developer pricing for frontier GPT-5.6 Sol model by more than 20%
- Alibaba releases Qwen3.8-27B open weights — a 27B multimodal model that beats Qwen3.7-Plus on several coding and agent benchmarks
- OpenCode Go quota exhausted in ~20 minutes after DeepSeek V4 Flash cache reads suddenly dropped to 0
- What locked-in price and usage limits would make you switch LLM providers?
Skills & plugins
- agent的安全相关skill 1. Trail of Bits Skills 一整套专业资安技能包，覆盖漏洞侦测、静态分析、完整审计流程，适合系统化做安全审查 https://github.com/trailofbits/skills 2. OpenAI Codex Secur
- Hi there! Yes, you can connect your own AI agent and add Binance Skills through Agent OS to access supported capabilities such as market dat
- feat: add CrewAI HOL Guard pre-tool runtime adapter
- Agentic provenance: framework plugins, AI-coding-harness capture, and provenance analysis
Repos, releases & updates
- kagura-agent/agent-id (0 stars) - 7 open issues
- Langfuse CLI is at v1. built for coding agents and power users who prefer the command line to interact with Langfuse data. rewritten engine,
- Blocking prompt injection deterministically costs 40% false positives
- OpenBot Review: Open Source AI Coworkers With a Commercial Catch
Funding & market
- AI chip startup Etched raises $700M at a $21B valuation — is AI inference the next big infrastructure battle?
- Nvidia to Pay AI Startup Poolside a $6B License, Newcomer Says
- [Live Webinar] Startup Funding Rounds in the AI Era
- SpaceX Attempted to Acquire AI Coding Startup Cognition