demiurge

Demiurge is an all-in-one AI skill that acts like a senior developer. It uses one strict rulebook to build secure apps from the ground up or completely fix broken ones.

The Agentic OS. One local screen over every AI agent you run: Claude Code + Codex, live, with receipts behind every "done". The multi-agent attention terminal.

Kill the AI slop, keep your voice. VERITAS, goddess of truth, strips em dashes, buzzwords, and AI tells, rewrites in YOUR voice, then audits its own rewrite. Claude Code skill.

Red-team your decisions. ATHENA runs a wise war council with a Sun Tzu terrain read: five adversaries, one GO/RESHAPE/KILL verdict on the record. Adversarial decision intelligence for Claude Code.

AI FinOps for AI agents. MONETA, goddess of the mint, stops your Claude Code agent burning tokens: warnings before waste, an honest lower-bound savings ledger. Token-discipline hooks.

The self-improving loop for AI agents. Chiron turns every correction into a permanent rule compiled into Claude Code, Cursor, and Windsurf. Correct once, never twice.

The memory layer for AI agents. HYPNOS, god of sleep, consolidates agent memory across CLAUDE.md, AGENTS.md, Cursor, Windsurf. Every change a diff, nothing ever deleted.

The house of gods that keeps AI agents honest: guardrails, AI FinOps, the memory layer, a self-improving loop, decision red-teaming, anti-slop, and agentic design. The forge is open, pick your god.

The sovereign AI foundation — identity, value, quality, and governance on the Demiurge chain. One package. Four primitives. Any application.

Guardrails for AI agents. HORKOS, god of oaths, checks the receipts: no proof, no "done". Evidence-audit hooks for Claude Code, zero LLM calls, deterministic benchmarks.

The agentic design agency in your terminal. The chief Muse runs the full engagement: interviews, signed briefs, role-agent production, a QA gate that rejects a 7.9. Directs Claude Design, Figma, v0.