Make AI coding agents architecture-aware: baseline-first, evidence-verified, drift-checked, and safe across long tasks.
Aegis is a method pack that guides coding agents to inspect a project's baseline, make bounded changes, and verify their work with fresh evidence. It is for people using coding-agent hosts who want fewer unverified changes and less unnecessary process. The catalogue add-ons implement this method through skills, instructions, commands, a hook, and a plugin.
Latest release v2.13.1 — v2.13.1: installation checks and host status / 安装检查与宿主状态 · 18 Oct 2026
3 files for GitHub Copilot, Codex, OpenCode and Claude Code: Aegis copilot-instructions.md, Aegis AGENTS.md, Aegis CLAUDE.md — 2,954 tokens loaded in every session.
.github/copilot-instructions.md A 149 tok AGENTS.md A 1,342 tok CLAUDE.md A 1,463 tok These files are GanyuanRan/Aegis's own configuration — they tell GitHub Copilot, Codex, OpenCode and Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.