Mechanism
In a multi-agent system, the conventional way to make several agents act together is a centralized scheduler or orchestrator that determines which agent executes which task, in what order, and with what resource allocation. The disclosure replaces that scheduler with forecasting-driven branch promotion. Each agent constructs its own planning graph from its own state, intent, and capabilities, evaluates its own branches through its own forecasting execution cycle, and begins executing a task not because a scheduler assigned it but because the agent's own forecasting engine generated a branch representing the task, the branch was classified as eligible, and the branch was promoted through the governance-validated promotion interface.
The coordination primitive is therefore the promotion of a branch from speculative to verified status. When an agent's forecasting engine promotes a branch, the promotion constitutes a self-directed execution commitment: the agent has determined, through its own cognitive evaluation, that the branch represents a viable, slope-eligible, policy-compatible, positively reinforced future state, and the agent commits to realizing that state through governed execution. Coordination emerges from the alignment and conflict resolution of independently generated plans rather than from a centralized authority that imposes plans from above.
The Planning Graph and the Promotion Interface
A planning graph is a mutable, memory-referenced, directed semantic structure that represents one or more hypothetical future states of the agent, its environment, or both. It comprises a root node representing the agent's current verified state and a plurality of branches, each branch representing a distinct hypothetical trajectory: a sequence of speculative mutations, delegation outcomes, environmental transitions, or intent resolutions. The planning graph is not an execution plan, a schedule, or a commitment; it is a pre-execution construct held in a computational domain that is structurally separated from the agent's verified execution memory.
No planning graph branch can directly modify verified execution memory. The boundary between the two domains is a promotion interface: a governance-controlled gateway that receives a candidate branch, subjects it to the full governance evaluation pipeline (policy compliance, trust slope validation, integrity impact assessment, capability verification), and either admits the candidate to verified execution memory as a committed mutation or rejects it and returns it to the speculative domain with a rejection annotation. The promotion interface is the sole gateway from speculative to verified status, and its governance requirements are not waivable by the agent's affective state, personality configuration, or operational urgency. Because every agent reaches execution only through this same gateway, branch promotion is the mechanism that replaces orchestration.
The Executive Engine and Branch Intersections
When multiple agents operate within a shared scope, such as a zone, a delegation hierarchy, or a coordination group, an executive engine aggregates their planning graphs into a unified executive graph representing the collective speculative state of the multi-agent system. The disclosure distinguishes two structural tiers: micro-planning graphs, which are the agent-level planning graphs each agent constructs and maintains from its own state, intent, and capabilities; and macro executive graphs, which are zone-level or group-level structures the executive engine constructs by aggregating, aligning, and reconciling the micro-planning graphs of all agents within its scope. The executive graph is not a simple union of agent-level planning graphs; it is a synthesized structure that identifies inter-agent dependencies, resource contention points, scheduling constraints, and cooperative opportunities not visible from any single agent's planning perspective.
The executive engine constructs the executive graph by collecting the current planning graphs from all agents within its scope, reading each agent's active branches along with their classification labels, affective reinforcement tags, slope projections, and policy compatibility flags. It then identifies branch intersections: pairs or groups of branches from different agents that reference the same environmental resources, target the same delegation endpoints, or project outcomes that depend on the actions of other agents. Branch intersections are the structural basis for coordination: they mark where agents' plans interact and where coordination, conflict resolution, or resource arbitration is required. The executive engine constructs executive graph nodes representing the coordinated state transitions required at those intersections, then evaluates the executive graph for global consistency, verifying that the coordinated plan does not produce trust slope violations at the zone level, does not exceed aggregate resource budgets, and does not violate zone-level policy constraints.
Arbitration Among Independently Generated Plans
The executive graph arbitrates among agents' planning graphs based on three criteria applied in priority order: slope compatibility, emotional reinforcement alignment, and personality profile alignment. Slope compatibility is evaluated first: branch combinations that maintain trust slope continuity at both the agent level and the zone level are preferred over combinations that maintain agent-level continuity but produce zone-level discontinuity. Emotional reinforcement alignment is evaluated second: branch combinations that produce positive affective reinforcement for the majority of participating agents are preferred over combinations that reinforce some agents but oppose others, subject to policy-defined minimum participation thresholds. Personality profile alignment is evaluated third: branch combinations consistent with each participating agent's personality field are preferred over combinations that would require agents to operate outside their personality-defined operating ranges.
The executive engine's role is to ensure that independently promoted branches across multiple agents do not conflict, not to determine which branches should be promoted. That determination remains with each agent's own forecasting engine. The executive graph maintains its own containment layer, structurally separate from the containment layers of the individual agents' micro-planning graphs, so that zone-level speculative coordination does not contaminate zone-level verified state, and so that promotion of executive graph branches to zone-level execution proceeds through its own governance-validated promotion interface.
Conflict Resolution When Branches Collide
When the executive engine identifies branch intersections in which agents' branches make conflicting demands on shared resources, project contradictory environmental outcomes, or require mutually exclusive execution sequences, it initiates a structured, deterministic conflict resolution protocol governed by policy-defined arbitration rules rather than ad-hoc negotiation or priority-based preemption. Overlap detection first classifies the conflict by type: resource contention, where multiple branches require the same finite resource at the same time; outcome contradiction, where branches project mutually exclusive environmental states; sequencing incompatibility, where branches require execution orders that cannot be simultaneously satisfied; and delegation collision, where branches target the same delegation endpoint with incompatible requests.
Compatibility assessment then evaluates whether the conflict can be resolved through branch modification, adjusting the timing, resource allocation, or execution sequence of one or more conflicting branches without changing their projected outcomes, producing one of three results: the conflict is resolvable through modification, it requires one or more branches to be suppressed or rerouted, or it is irreconcilable and requires escalation to governance authorities. When a conflict cannot be resolved through modification, arbitration selects which branch takes precedence using the three-criteria priority ordering, additionally considering each branch's integrity impact, the hierarchical position of each agent in the delegation hierarchy, and a global impact assessment that prefers branches whose execution benefits a larger proportion of the agent population. The disclosure also describes an emotional quorum override that acts only as a tiebreaker within governance constraints when standard arbitration is inconclusive, and personality-driven rerouting that, for an agent whose personality field indicates low conflict tolerance or high fallback rigidity, replaces a conflicting branch with an alternative branch from the same agent's planning graph rather than suppressing it outright.
Cross-Agent Planning Graph Visibility
In coordination contexts, an agent may selectively expose portions of its planning graph to trusted peer agents through a policy-governed visibility interface. The exposed portions are read-only copies of selected branches, transmitted with their speculative markers intact, enabling peer agents to observe the exposing agent's speculative landscape without contaminating their own verified execution memory. The exposing agent's policy configuration specifies which branches may be exposed, which peer agents are authorized to receive them, the maximum exposure depth, and the exposure duration.
Cross-agent visibility enables coordinated speculative reasoning without centralizing planning authority: peer agents can align their own planning graph construction with the exposed landscape, identify complementary branches, avoid redundant speculation, and detect potential conflicts between their respective projected futures. Each exposure event is recorded in the exposing agent's lineage, and the receiving agent's lineage records receipt of the exposed speculative content with the appropriate speculative marker, preventing inadvertent contamination of verified state.
Architectural Consequences
Replacing the scheduler with forecasting-driven branch promotion eliminates the single point of failure: if one agent's forecasting engine fails, other agents continue to construct, evaluate, and promote their own branches. It distributes the computational burden of planning, because each agent's forecasting engine operates on the agent's own state rather than requiring a centralized scheduler to model all agents simultaneously. It preserves agent autonomy, because each agent's planning is shaped by its own personality, affective state, integrity field, and policy constraints, producing plans structurally aligned with the agent's individual characteristics. And it supports heterogeneous agent populations: agents with different personality configurations, different capability envelopes, and different policy constraints can coexist and coordinate without a centralized scheduler that must model their differences.
The disclosure further describes deployment topologies for this coordination model. In a centralized deployment, the forecasting and executive engines run on a single node or cluster with planning graphs partitioned by agent identity. In a federated deployment, forecasting engines run at individual agent nodes while the executive engine runs at zone-level aggregation nodes and collects planning graph summaries rather than full structures. In a decentralized deployment, both forecasting and executive engines run at individual agent nodes and executive graph construction proceeds through peer-to-peer coordination, eliminating the zone-level executive engine as a centralized coordination point. The architectural properties of branch promotion, the promotion interface, the containment layer, and executive graph aggregation are invariant across these deployments; the deployment model affects communication topology, latency, and resource allocation but not the governance requirements.
Disclosure Scope
The forecasting-as-coordination-primitive architecture, comprising per-agent planning graphs constructed and evaluated by each agent's own forecasting engine, the promotion interface as the sole governance-validated gateway from speculative to verified status, the executive engine that aggregates micro-planning graphs into a macro executive graph and detects branch intersections, the three-criteria arbitration ordering of slope compatibility, emotional reinforcement alignment, and personality profile alignment, the structured conflict resolution protocol of overlap detection, compatibility assessment, and arbitration with its emotional quorum override and personality-driven rerouting, and the policy-governed cross-agent planning graph visibility interface, is disclosed in the cognition filing (U.S. Application No. 19/647,395 and its international counterpart). This article describes that disclosed mechanism. The scope extends to the centralized, federated, and decentralized deployment topologies described, provided coordination continues to emerge from the alignment and conflict resolution of independently promoted branches rather than from a centralized scheduler that imposes plans on the agents.
