<aside>
This document describes the cognitive architecture underlying Cognis, a neuro-symbolic Society of mind designed for general-purpose intelligence across agentic and physical domains. The architecture is intentionally domain-agnostic: the same structural principles apply whether the system is orchestrating software agents, controlling a robotic arm, or coordinating a multi-step workflow.
</aside>
Public research note · Onzore
The thesis is simple: intelligence becomes more powerful when action, correction, and failure turn into reusable capability.
Thesis
AI is moving from systems that answer to systems that act. Once intelligence starts acting, the bottleneck changes. It is no longer enough for a model to produce a good next sentence or a good plan. The system has to preserve state, choose the right capability, verify the action while it is happening, recover when the world changes, and make the next attempt better than the last one.
That is the direction behind Cognis.
Cognis is the brain that lets limited capabilities become coherent action. It works below the level of a single process or a single agent. The system does not only route a request. It decomposes capability, recombines useful pathways, evaluates execution, and strengthens the traces that worked.
Because physical machines act inside changing reality, this matters directly for Physical AI. A robot does not need only a better instruction. It needs a way to turn action, correction, and failure into reusable capability.
Cognis is the foundational intelligence layer on which harnesses, agents, and intelligent systems are built. It provides the structure that determines how any model sees context, executes through tools, gets evaluated, and retains what it learns over time. These four functions — context, execution, control, and accumulation — are the infrastructure beneath model intelligence.
In mid-2025, we began building toward context graphs as one research direction for this architecture. The core observation was that most agent frameworks treat orchestration as a routing problem: a query arrives, gets classified, and gets dispatched to a tool or model. That pattern handles single-turn tasks. It fails when a request crosses domains, depends on learned preferences, or requires adaptation mid-execution. Orchestration coordinates. It does not learn. It does not decompose.
Cognis operates below the level of the process. It works with primitives, what we call sparks, and the learned connections between them. A spark is a unit of capability at any level of abstraction: it can be an atomic function, a composed workflow, or a full agent. The system evaluates spark combinations through lightweight critics and learns execution patterns over time through weighted connections we call K-lines. The result is a system that does not reason from scratch on every request.
The architecture draws from Marvin Minsky's Society of Mind, specifically the insight that intelligence is not a single process but an emergent property of many simple components interacting. We extend Minsky's framework in several ways. K-lines carry outcome-based weights rather than binary activations. Critics both excite and inhibit rather than only suppressing. The abstraction boundary between node and connection is intentionally fluid. A K-line that fires reliably as a unit is, to the layer above it, indistinguishable from a spark. An agent is a node with its own internal K-lines. The hierarchy is recursive.
Consider a request: "I want spicy pasta and book a cab."