Aura is a self-hosted, multi-user AI agent platform written in Go that gives each person a distinct identity, memory database and sandboxed workspace. It runs as a Docker Compose appliance (Postgres, ArcadeDB, Garage S3 object store plus sidecars) and exposes a web cockpit, CLI, and channels like Telegram and WhatsApp. Memory is graph-native (ArcadeDB) with temporal facts, provenance and per-identity isolation; conversations live in Postgres. Models are provider-neutral and hot-swappable from the cockpit (OpenRouter default, ChatGPT plans, local llama.cpp or Ollama), so inference can run through cloud endpoints or a local server. Hardware requirements are modest (a 16 GB mini PC is enough for the default stack), and installation is automated via an npx create-aura-appliance installer.
Functionally, Aura implements a streaming agent loop with budgeted steps, deferred tools and on-demand tool schema loading, an adaptive reasoning router that paces effort by model capabilities, document retrieval with source citations, and a scheduler for reminders and agent jobs. Operators get per-identity sandboxes (gVisor optional), fine-grained capability grants, audit views and TOTP-enabled sign-in via Authula. The cockpit includes media studio features, live steering/approvals, and bundled integrations (calendar/email PIM, SearXNG search, WhatsApp, Cloudflare). The stack emphasizes test coverage, CI checks, and distribution as a Docker Compose appliance under an MIT license.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.
Four coding agents were given a $100 budget to build a functional PDF editor, but their implementations had noticeable bugs and shortcomings. The experiment highlights that current AI agents struggle with goal pursuit and user interaction, leading to subpar software quality. (blog.nielstron.de)
Lithos-ai open-sourced lithos-metal, enabling Qwen3.8-27B to run at over 200 tokens per second on an Apple M5 Max laptop. This allows ultra-fast inference for coding agents with a single command. (twitter.com)
Google Cloud introduces the Gemini agent, a universal AI tool that handles questions, content creation, coding, and knowledge work within familiar applications like Gmail and Docs. It features industry-specific skills, security, cost controls, and integrates seamlessly into existing workflows. (cloud.google.com)
Temporal developed pi-temporal to enable the Pi coding agent to recover from machine failures during tasks by leveraging a fleet of workers and durable execution. This setup allows seamless task continuation across machine failures without repeating interrupted commands. (temporal.io)
![OpenAI cannot make AI safe on its own [pdf]](https://hntoday.pixelshot.io/screenshots/20261008/0217d5d1410143c5_tn.jpeg)
OpenAI has acknowledged that it cannot ensure the safety of artificial intelligence systems on its own. The document discusses the challenges and limitations faced by OpenAI in making AI safe without external collaboration or oversight. (mikitabalesni.com)
Yoshua Bengio urges AI researchers at frontier companies to prioritize safety and consider leaving if safety measures are insufficient. He warns that current development races risk pushing humanity toward catastrophic outcomes and calls for slowing AI progress to prevent loss of control. (transformernews.ai)
Today's best Hacker News stories, summarized and screenshotted, one email a day.