GoKawiil
Tech News
LayerFS is a filesystem designed for AI coding agents that automatically creates a checkpoint at every step an agent takes. Each file operation is recorded as a commit, and each Bash command commits once it finishes executing. Calls that make no changes do not generate a commit, avoiding unnecessary history noise.
The developer behind Pi has released version 1.0 with a feature called Codemode, which brings Model Context Protocol (MCP) support into the harness. The post explains how tool-calling works for LLMs, why bash/CLI composition has been favored, and where that approach breaks down for things like image reading or spawning sub-agents.
Geocodio, a small geocoding API company, describes evolving from scattered bash scripts and Laravel commands into full internal applications with mobile-friendly interfaces and strong test coverage. The company credits recent AI coding tools with enabling this jump in ambition for building software to support its own operations, including an existing custom deployment visualization tool.
The author recounts struggling for years to move beyond single terminal commands, unable to chain logic like conditionals, loops, or piping output between programs. Over time they gradually learned shell scripting well enough to write basic conditional scripts, such as one checking npm install exit codes and printing custom error messages.
A security researcher's report describes Omarchy 4.0, the Linux distribution promoted by DHH, as riddled with basic vulnerabilities, including bash injection triggered by video titles and notifications capable of executing arbitrary commands. The report argues these are not obscure edge cases but well-known classes of input-handling flaws that mature software practices routinely prevent, and suggests some scripts may have been AI-generated without proper review.
Laude has open-sourced Headlong, an agent framework built around a Bash core under 10,000 lines that keeps AI agents continuously generating thoughts rather than only responding to prompts. Instead of waking on a schedule to run a checklist, a Headlong agent runs a persistent thought stream where incoming messages are treated as observations, and it decides for itself whether and when to reply. Laude tested this internally with a shared agent named Audel across Slack, Telegram, and mobile, letting multiple team members interact with the same continuously running agent.