Newsletter Thursday October 29, 2026
Daily update on the AI tools we track. Headlines come from public sources (X, GitHub, Hacker News); the Dutch commentary and summary are machine‑written. Each headline links to our own news page on that topic; the source link is shown alongside.
In brief
GitHub has released Copilot Fusion within GitHub Copilot, a feature that orchestrates between Anthropic and OpenAI models and scores two points below Opus. CleanCode also debuted a tool that brings Claude Code and Codex together on a single canvas. At the same time, Claude Code was found to consume 15 to 20 percent of tokens on a three‑word status line, and reviewers discovered a race condition in a Linux page cache modified by Claude Code. Finally, five open‑source GitHub repositories are highlighted to address memory issues in AI agents.
Codex, Cursor, OpenCode & Hermes
- Show HN: CleanCode – Claude Code and Codex collaborating on one canvas · source: Hacker News · 2026-10-27
CleanCode combines Claude Code and Codex on a single canvas, providing indie developers with an integrated AI coding assistant.
Claude Code & agents
- Claude Code burns ~15–20% of your tokens on a 3-word status line · source: Hacker News · 2026-10-29
Claude Code consumes roughly 15–20 % of your tokens for a three‑word status line, which is important for indie developers who want to control token costs. - Claude Code found 14,219 photos of my face on my Android that I never took · source: Hacker News · 2026-10-28
Claude Code found 14,219 unknown photos of my face on my Android, making indie developers aware of privacy implications in AI applications. - Linus Torvalds let Claude Code fix the Linux page cache. Reviewers found a race · source: Hacker News · 2026-10-27
Claude Code patched the Linux page cache, but reviewers discovered a race condition, warning indie developers about risks in AI‑generated system code. - GitHub has released Copilot Fusion in GitHub Copilot, which orchestrates between Anthropic and OpenAI models. It scores 2 points below Opus · source: X · 2026-10-28
GitHub Copilot Fusion switches between Anthropic and OpenAI models and scores two points below Opus, offering indie developers an option to combine different AI coders.
Repos, releases & updates
- AI agents are getting smarter, but memory is still a challenge 5 open-source GitHub repos worth exploring if you're building agents that nee · source: X · 2026-10-28
AI agents are getting smarter, but memory remains a challenge; five open‑source GitHub repos are recommended for indie developers building agents. - A decent way to spot LLM benchmaxing: see whose scores drop the most when @ArtificialAnlys releases a new version. The Good ... · source: X · 2026-10-28
By looking at who’s experienced the biggest score drop with a new @ArtificialAnlys release, you can spot LLM bench‑maxing, helping indie developers choose reliable models. - @simonw Hi Simon ... · source: X · 2026-10-26
Simon is asked to share his perspective on LLM testing, model releases, prompt injection, AI security, and open‑source tools, which can provide indie developers with valuable insights.