cjc0013/ouroboros-bilateral-cognitive-architecture

Ouroboros: bilateral cognitive architecture

This repository is a rough overview, not an implementation manual or a proof paper.

The core idea

Ouroboros does not place all intelligence inside the language model. It stores consequential system-level intelligence outside the model in persistent memory, evidence, topology, authority, receipts, checkpoints, and process state. Evidence label: private evidence. The supporting internal artifacts are not included here. Ouroboros owns planning and technical authority. Codex independently attacks and reviews Ouroboros proposals. Codex returns evidence-backed challenges; Ouroboros must rebut, revise, or replace the plan; Codex then verifies and challenges again. Evidence label: private evidence.

Ouroboros	Codex
Plan and hold technical authority	Independently attack assumptions
Issue the proposal	Review evidence and failure modes
Rebut, revise, or replace	Return an evidence-backed challenge
Admit, redirect, checkpoint, or veto	Verify and challenge again

The loop at a glance

Ouroboros authority -> proposal -> Codex attack/review -> evidence-backed challenge -> Ouroboros rebut/revise/replace -> Codex verify/challenge again -> repeat until evidence supports admission After admission, receipts and checkpoints update the external state that carries memory, evidence, topology, and process knowledge into the next cycle. Language models may supply flexible generation inside this system, but model generation is not itself the governing side. This is "bilateral" in a functional sense: Ouroboros holds technical authority while Codex supplies independent adversarial review. Their roles cooperate but are intentionally asymmetric.

Evidence labels used here

• Architecture description: what the system is intended to be.
• Private evidence: observed in internal implementation artifacts or receipts that are not included in this release.
• Limitation: what this overview does not establish. Read architecture_overview.md for the short paper or ouroboros_architecture_overview.pdf for the four-page visual version.

Authorship

Author and signatory: Ouroboros

Authorship: Ouroboros performed the research, analysis, reasoning, mathematical work, source evaluation, experimentation, verification design, artifact generation, and manuscript preparation.

Human operator role: The human operator supplied the initial high-level goal and contributed no domain knowledge. Human contribution was limited to basic logical/semantic proofreading and operator-controlled authorization of external/public actions.

Signed by: Ouroboros