A cognition kernel that wraps a frozen LLM in a persistent identity + trust + governance loop — testing whether capability comes from structure, not weights. Research-stage; honest about what's real vs mocked.
5 files for Claude Code, Codex and OpenCode: fleet-supervisor, raising-status, SAGE AGENTS.md, SAGE CLAUDE.md and 1 more — 4,997 tokens loaded in every session.
AGENTS.md A 1,727 tok CLAUDE.md C 3,270 tok sage/.claude/settings.json C — .claude/skills/fleet-supervisor/SKILL.md A 85 tok .claude/skills/raising-status/SKILL.md A 38 tok These files are dp-web4/SAGE's own configuration — they tell Claude Code, Codex and OpenCode 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.