A self-evolving Agent OS — every interaction sharpens how it judges, not just what it knows. Four ideas: self-evolution (mistakes become code gates, not logged lessons), brain-first domain brains that grow and decay, cognition that persists between sessions, and proprioception — it senses and drives its own body. Human directs. AI delivers.
Latest release v2.3.0 — SwarmAI v2.3.0 · 15 Oct 2026
11 files for Claude Code, Codex, OpenCode and Kiro: PreToolUse, SwarmAI, SwarmAI AGENTS.md, SwarmAI CLAUDE.md and 7 more — 17,424 tokens loaded in every session.
AGENTS.md C 5,669 tok CLAUDE.md A 355 tok .kiro/steering/context-and-memory-safety.md A 1,349 tok .kiro/steering/debugging-methodology.md A 459 tok .kiro/steering/multi-tab-isolation-principles.md A 2,881 tok .kiro/steering/security-coding-baseline.md D 1,858 tok .kiro/steering/self-evolution-guardrails.md A 1,475 tok .kiro/steering/session-identity-and-backend-isolation.md A 1,706 tok .kiro/steering/swarmai-dev-rules.md A 1,672 tok .claude/settings.json A — .claude/settings.json A — These files are xg-gh-25/SwarmAI's own configuration — they tell Claude Code, Codex, OpenCode and Kiro 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.