The Agent! Blog
Straight from the codebase: how AgentiLoop Agent! works, what changed, and why. New posts daily.
Why Agent! Compacts at Half the Window, and the Three Bugs That Shrank It to 16K
How Agent!'s context compaction decides when to summarize a long task, and the September 28 fixes for fallback models, Ollama and a 272K Codex window.
A Release on the First of Every Month, and Pre-Releases in Between
Agent! now ships a stable release on the 1st of every month. Daily pre-releases (moving to weekly) are where new features and fixes get tried out, and release candidates lock things down before the big day.
The Terminal Strikes Back: Taking the Agent Loop Cross-Platform in Rust and Go
Agent! is a Mac app, but a coding agent belongs wherever developers work. Here's how the agent loop became two CLIs, one in Rust and one in Go, with a full-screen TUI on macOS, Windows and Linux.
Agent! Now Runs on macOS 14.6 and Intel Macs: How We Got Off macOS 26
Agent! was built around Apple Intelligence on macOS 26. Here's how one day of #available gates and package bumps brought it to macOS 14.6 on Apple Silicon and Intel.
Before the Model Gets a Vote: Inside Agent!'s Shell Guardrails
How ShellSafetyService, the daemon-side re-check and the Jev second opinion stop an AI agent from wiping your Mac, walked through the actual Swift code.
Welcome to the Agent! Blog: One App, Any AI, Total Command Over Your Mac
What AgentiLoop Agent! is, how it's built, and what this daily blog will cover, straight from the source code.
Meet Jev: A Second Opinion That Answers in Types, Not Prose
Agent! now asks Jev (TypeSafe System One) how destructive a shell command is before running it. Here's why Jev is an advisor and not another LLM provider.
Why We Ripped Out Coding Mode: Let the Model Pick Its Own Tools
Agent! used to guess whether you wanted coding or automation and hide tools to match. One broken Photo Booth task showed why the harness should stop second-guessing the model.
