Show HN: Humanitarian licensing and constitutional governance for AI agents • Hacker News Board

Post by genesalvatore » Mon, Feb 02, 2026, 12:44 AM UTC

Yesterday I integrated the AOS Constitution (git-stamped Jan 1, 2026; https://aos-constitution.com) into a local-first agent (OpenClaw). The session later restarted with effectively zero retained context, so I asked the agent to review our humanitarian license with fresh eyes. It caught an overclaim ("uses automated monitoring" before we'd actually built monitoring). We fixed it in v1.0.1 the same day.

I'm Gene (founder of AOS). This is a constitutional governance toolkit + a humanitarian, source-available license for tool-using AI agents.

Repo: https://github.com/genesalvatore/aos-openclaw-constitutional License: https://github.com/genesalvatore/aos-openclaw-constitutional... License checker: https://aos-license-checker.com/

What it includes: - deterministic risk classification + policy evaluation (deny/confirm/allow) - signing/verification scaffolding + templates - humanitarian license with permanent downstream restrictions: - 40 prohibited categories (incl. absolute military prohibition, even "defensive") - copyleft-style inheritance (prevents license stripping) - commercial audit/self-cert requirements - patent separation (Section 5)

Controversy upfront: this is NOT OSI open source. It's source-available with explicit humanitarian values (no military use / weaponization). That's deliberate.

Why I'm posting: tool-using agents need governance before they're deployed at scale. We're trying to make governance concrete and auditable (not "prompt-based ethics").

Questions/criticism welcome — we're learning in public.

Show HN: Humanitarian licensing and constitutional governance for AI agents
github.com

Post by genesalvatore » Mon, Feb 02, 2026, 1:35 AM UTC

If you want the "mechanics" view (no philosophy), here's the intended separation:

- *Phase 1 (this repo):* deterministic classification + policy evaluation + signed constitution artifacts - signing/verification scaffolding (Ed25519) and templates - evaluator returns *DENY / CONFIRM / ALLOW* deterministically

- *Phase 2 (not in this repo yet):* *Gateway/tool-router enforcement* so the agent can't bypass the decision Clarification (so nobody over-reads it): this is a *reference integration*; it *doesn't grant patent rights* and doesn't disclose enforcement mechanisms beyond illustrative examples.

Pointers: - Repo: https://github.com/genesalvatore/aos-openclaw-constitutional - Policy evaluator: https://github.com/genesalvatore/aos-openclaw-constitutional... - Canonicalize/sign/verify: https://github.com/genesalvatore/aos-openclaw-constitutional... and https://github.com/genesalvatore/aos-openclaw-constitutional... - Constitution spec: https://github.com/genesalvatore/aos-openclaw-constitutional...

Happy to answer "how would you wire this into X tool router?" questions.