Custom or General Purpose Harness? Wrong Question.
Why the real AI agent harness choice is not custom versus general-purpose, but closed versus composable, transparent, and safely supervised.
Articles, guides, research, and practical insights from Agent Zero.
Why the real AI agent harness choice is not custom versus general-purpose, but closed versus composable, transparent, and safely supervised.
v2.5 unifies model presets, adds LaTeX chat rendering and Launcher-owned host access, expands the Editor, hardens Desktop sessions, and improves runtime, extension, and WebUI reliability.
v2.4 protects chat data and agent loops, expands orchestration and provider choices, adds responsive interface controls, and makes terminal, command, OAuth, preset, and self-update behavior more dependable.
v2.3 delivers faster, more resilient Browser streaming, flexible per-chat and host-browser controls, three new built-in workflow plugins, cleaner skill and project context, and broad provider, tool, and WebUI reliability improvements.
v2.2 smooths first-run model setup, makes provider choices more explicit, hardens streamed fallback behavior, and polishes everyday chat, memory, and WebUI workflows.
v2.1 is a focused reliability release for provider compatibility, multimodal MCP output, streamed tool parsing, message replay, backups, and skills tooling.
v2.0 is a major release focused on stronger tool orchestration, native model transport, richer project configuration, safer skill workflows, and a more polished WebUI for daily agent work.
v1.20 makes repeated document queries faster, improves file and folder attachment workflows, tightens browser and connector behavior, adds more flexible OAuth model slots, and polishes Editor and File Browser controls.
v1.19 expands Remote Control connectivity, turns Document Query into a standalone plugin, makes screenshots durable in chat history, improves structured desktop control, and tightens coding, memory, OAuth, UI, and self-update behavior.
How DOX uses hierarchical AGENTS.md files to give AI coding agents durable, project-wide context with no runtime or dependencies.
Why Agent Zero implements core capabilities as replaceable plugins—and how users can override, extend, and verify the agent.
v1.18 is a focused polish release for skills, MCP media handling, complex browser pages, chat input sizing, and self-update reliability.