zerostack

You will notice this one by how little you notice it. The trait that decides it in daily use is restraint with your machine: a compact binary and a low memory footprint mean it sits beside a language server and a build without turning your laptop into a heater, which sounds trivial until you have three sessions open. Sessions save and resume, so closing the terminal is not a decision.

Pick it if lightness is what you have been missing. Pick OpenCode when you want a bigger community around the same design, or a desktop tool if you want to see everything at once.

The default is where this breaks. Protection relies on a platform-specific backend, and if that backend is missing the run continues anyway rather than stopping, so an engineer who asked for containment gets the appearance of it. There is a second flag that makes it required, and needing a flag to make a safety feature actually safe is a design decision. The worktree and loop features are labelled experimental on top of that.

What it does right: five permission modes with per-tool glob rules, which is a genuinely granular policy surface rather than a single trust switch.

The architecture addresses the failure modes that actually end long sessions. 1. Context is compacted automatically as it grows, so degradation from an overfull window is handled by the scaffold rather than by the user noticing. 2. Repetition is detected and interrupted, which targets the loop where an agent alternates between two wrong edits indefinitely. Both are bounding conditions, and bounding conditions are what separate a harness from a chat client.

No evaluation accompanies either claim, and the iterative mode is labelled experimental by its own author, which is the appropriate epistemic status for a mechanism this hard to test.

This is a personal project of high quality with no commercial apparatus around it at all: no entity, no tier, no hosted service. Copyleft closes the usual escape route where a maintainer later sells a proprietary edition, so even a change of heart has limited options. The likely trajectory is enthusiast maintenance for as long as the author enjoys writing it.

A small share of attention in a category where the incumbents are also free means no leverage of any kind. No acquirer exists; the plausible exit is a maintainer job offer. Position: use it, contribute if you rely on it, and assume you are the support contract.

A support matrix I cannot control is worse than a missing feature. Memory, hooks and several protocol integrations are build-time options rather than runtime settings, so an engineer who installed from a package and one who built their own are running different capability sets under the same version number, and my helpdesk cannot tell them apart. Supported platforms are macOS and Linux only.

There is a headless flag, so a pipeline step is possible, and the row records no single sign-on, SCIM or audit trail. Onboarding is an hour. Approved with conditions: one blessed build, distributed by us.

Strong copyleft, which is the licence I want on something that touches my repository, because anyone who ships a modified version owes it back. Installation is a cargo command, so I build it from crates on whatever machine I like rather than trusting a binary someone else produced. The providers include a local runtime, an inference server and a routing library, so inference can stay in the room with me.

Rust means the fork is compilable and the memory behaviour is predictable. This is the sort of project I would send patches to instead of complaints.