MENTAT is a personal memory system designed to stop AI chats from repeatedly asking for the same background. It stores concise, linked Markdown notes in a private Git repository, organized around personal context, life areas, projects, people, and dated monthly logs. The owner edits the notes directly in Obsidian or another editor; wiki links provide structure, while automated builds create an index and graph. A Cloudflare Worker exposes the notes to Claude and OpenCode through MCP tools for reading, searching, filtering, following links, and retrieving recent log entries. Every push to the main branch rebuilds and deploys the Worker, avoiding a separate database and keeping changes synchronized across devices.
The system prioritizes privacy and controlled updates because AI-written notes could mislead future chats if untrusted content were recorded as fact. Access requires GitHub OAuth tied to the owner’s immutable numeric user ID; the Worker runs on a private subdomain, and the repository is checked for secrets. AI write requests create pull requests rather than editing trusted notes directly. The Worker restricts paths and content, log-entry requests can auto-merge after 72 quiet hours, and other note changes require approval. A graph page visualizes note connections behind the same sign-in. The author also describes practical implementation failures, including brittle date-based tests, OAuth callback confusion, and an auto-merge rule that failed on real changes. A public, fictional demo and reusable template show how others can start with a short profile note and a few project notes.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.
Richard Sutton argues that brute-force computation alone cannot produce intelligence and that AI needs clever new methods. He says the science of AI is far from finished. (twitter.com)
Anthropic is accused of charging Russian Claude users before identity checks, then cutting off access after verification and retaining biometric data. Critics say account bans can affect linked GitHub projects and Russian citizens living abroad, and accuse the company of failing to answer public questions. (twitter.com)
LLM coding agents boost workplace productivity but can prevent developers from learning through hands-on problem-solving and weaken their technical skills. Developers should preserve their expertise by building projects manually, studying fundamentals, and practicing without AI tools in their own time. (ducktape.blot.im)
A fabricated Newtype interview led The Diplomat to retract its article about a Korean startup’s AI war test in Ukraine. The retraction followed a diplomatic dispute. (thediplomat.com)
Microsoft Research’s AI Frontiers group brought together about 50 researchers to develop agent systems and foundational AI advances. After the team joined Microsoft AI in September, its research-first approach clashed with leadership’s emphasis on scaling. (hunch.net)
Microsoft researcher John Langford has left the company. Microsoft’s AI Frontiers group has been declared finished. (businessexaminer.ca)
Today's best Hacker News stories, summarized and screenshotted, one email a day.