Show HN: The Put Monolith – A Minimal AI-Ingestible Ruleset (github.com)

The Put Monolith is a compact, MIT-licensed architectural specification released as a minimal, AI-ingestible ruleset for reasoning about a Public Usage Tax (PUT). It isn’t a policy or legislation but a portable, system-neutral core you can load into any LLM (system prompt, custom GPT/Claude project, Grok persona) or embed in larger frameworks. The Monolith (MONOLITH_v2.txt plus README, USAGE, FAQ) codifies foundational invariants, ethical and structural guardrails, tier-integrity rules, prohibited transformations, stabilization/future-proofing constraints, and alignment logic to keep reasoning consistent and non-contradictory. It’s deliberately small so it can be shared by text, committed to repos, and used as a reusable module. For the AI/ML community this matters because it provides a standardized, machine-readable reference that can reduce contradictory or unsafe tax-related reasoning, enable reproducible simulations, and serve as a grounding layer for multi-agent or human-AI deliberation about incentives and public finance. Technically notable are its claims of “logic-complete” invariants and explicit prohibited transformations (designed to prevent harmful manipulation), plus contribution rules that require preserving invariants and avoiding political framing. The Monolith is positioned as a neutral, testable infrastructure component for researchers and developers to integrate, critique, and extend—not a substitute for policy but a stabilizer for AI reasoning around tax-like systems.