Mechanism
The forecasting execution cycle is the procedure by which the forecasting engine constructs, evaluates, and classifies the branches of a planning graph. It is invoked at each cognitive decision point, that is, whenever the agent must evaluate candidate actions, select among alternatives, or determine whether to act, delegate, or defer. The cycle is not a background process or a periodic batch operation. It is a synchronous component of the agent's deliberation pipeline that executes whenever the agent faces a decision requiring speculative evaluation, and it comprises six sequential phases: initialization, speculative mutation simulation, slope projection and validation, policy compatibility check, emotional reinforcement tagging, and branch marking and pruning.
Each phase operates on a planning graph, a mutable, memory-referenced, directed structure whose root node represents the agent's current verified state and whose branches each represent a distinct hypothetical trajectory. Throughout the cycle, the planning graph remains structurally separate from the agent's verified execution memory. No phase writes to verified state. The cycle produces classified branches, and only a branch promoted through the separate governance-validated promotion interface becomes a committed mutation.
Phase 1: Initialization
The forecasting engine reads the agent's current verified state, including the intent field, context block, memory field, affective state field, integrity field, and policy reference field, and constructs or updates a planning graph with a root node representing that current verified state. If a planning graph is already active for the current decision context, initialization refreshes the root node with current state and re-evaluates the existing branches for continued viability. If no planning graph exists, initialization creates a new one and generates an initial branch set using the engine's instantiation logic. The root node is a snapshot: it does not establish a live reference, so subsequent verified state changes do not automatically propagate into the graph, and the evaluation is deterministic with respect to the state captured at construction.
Phase 2: Speculative Mutation Simulation
For each active branch, the forecasting engine simulates the hypothetical mutations that the branch represents. The simulation applies the speculative mutations to a sandboxed copy of the agent's state, not to verified execution memory, and computes the projected outcome of each mutation sequence, including projected environmental responses, projected delegation outcomes where the branch involves delegation, and projected secondary effects on the affective state and integrity fields. The simulation is deterministic: given the same input state and mutation sequence, it produces the same projected outcome. This distinguishes the simulation from statistical tree search methods such as Monte Carlo Tree Search, because each step operates deterministically on defined structural fields, produces reproducible outcomes, and is constrained by trust slope continuity and policy compatibility at every step rather than being evaluated statistically over random rollouts.
Phase 3: Slope Projection and Validation
For each simulated branch, the slope validation module computes a hypothetical Derived Anchor Hash, denoted DAH-prime, by applying the branch's speculative mutation sequence to a sandboxed copy of the agent's lineage and computing the trust slope hash that would result. The hypothetical DAH-prime is compared against the agent's current trust slope trajectory using the same continuity validation algorithm that the governance infrastructure applies to committed mutations. A branch whose DAH-prime maintains continuity is slope-eligible. A branch whose DAH-prime breaks continuity, by producing a lineage gap, a hash chain discontinuity, or a provenance violation, is slope-ineligible. The phase also computes the magnitude of the trust slope continuation, how far the branch advances or retreats along the trajectory, enabling comparative ranking of slope-eligible branches. This filtering is prospective: it removes ineligible branches before they reach the promotion interface, so the governance pipeline never receives a candidate that would fail trust slope validation.
Phase 4: Policy Compatibility Check
Each slope-eligible branch is evaluated against the agent's current policy configuration to determine whether its speculative mutations are admissible under the agent's governance constraints. Branches that contain mutations excluded by policy are flagged as policy-incompatible and reclassified or pruned. Branches that satisfy all policy requirements are confirmed as policy-compatible. The policy check is applied only to branches that have already passed slope validation, concentrating the engine's evaluation on branches that retain a viable path to execution.
Phase 5: Emotional Reinforcement Tagging
Each slope-eligible, policy-compatible branch receives an affective reinforcement tag computed by the affective prioritization module. The tag encodes the emotional valence of the branch, the degree to which its projected outcome aligns with the agent's current affective disposition, and it influences the branch's priority in subsequent evaluation and promotion decisions. Branches with strong positive reinforcement are prioritized for promotion. Branches with strong negative reinforcement are deprioritized but retained for introspective analysis. The affective modulation shapes how branches are constructed, prioritized, and retained, but it does not determine admissibility: the governance requirements for promotion remain identical regardless of the agent's affective state.
Phase 6: Branch Marking and Pruning
Following the evaluation phases, each branch receives a classification label. A branch that is slope-eligible, policy-compatible, and positively or neutrally reinforced is marked eligible, a viable candidate for promotion. A branch that is slope-eligible and policy-compatible but negatively reinforced is marked introspective and retained for cognitive self-examination rather than promotion. A branch that is slope-eligible, policy-compatible, and suited for transfer to a child agent is marked delegable. A branch that fails slope validation, policy compatibility, or both is marked pruned and scheduled for removal by the pruning manager. Classification is not permanent: the cycle re-evaluates branch classifications at each iteration, so an introspective branch may become eligible when the affective state shifts, an eligible branch may become pruned when supporting conditions change, and a delegable branch may become eligible when a delegation target is unavailable.
Composition
The execution cycle is the operative loop of the forecasting engine, whose five principal components correspond to the cycle's work: instantiation logic generates the initial branch set in initialization; the affective prioritization module supplies the reinforcement tags in phase five; the slope validation module performs phase three and also flags slope-ineligible branches; the personality-based modulation filter shapes the breadth, depth, risk profile, and temporal horizon of branch construction and evaluation; and the pruning manager carries out the removals scheduled in phase six. The cycle re-runs slope invalidation pruning at each iteration, so a committed mutation that changes the trust slope trajectory can render a previously slope-eligible branch discontinuous and reclassify it as pruned.
Because the cycle terminates in classified branches rather than committed actions, it composes with the governance-validated promotion interface, which is the sole gateway from speculative to verified status and subjects each candidate to the full governance evaluation pipeline. An eligible branch begins executing not because a scheduler assigned a task but because the agent's own cycle generated the branch, evaluated it as eligible, and the branch was promoted through that interface. Where multiple agents share a scope, the executive engine aggregates their planning graphs and resolves branch intersections, so coordination emerges from independently generated and independently classified plans rather than from a centralized authority.
Disclosure Scope
The forecasting execution cycle, comprising the six sequential phases of initialization, speculative mutation simulation, slope projection and validation, policy compatibility check, emotional reinforcement tagging, and branch marking and pruning, invoked synchronously at each cognitive decision point, operating on a planning graph held in structural separation from verified execution memory, and producing branches classified as eligible, introspective, delegable, or pruned, is disclosed in the cognition filing (U.S. Application No. 19/647,395 and its international counterpart) at Section 4.4. This article describes that disclosed mechanism. The scope extends to the deterministic, slope-constrained character of the speculative simulation, to the prospective filtering that prevents slope-ineligible or policy-incompatible branches from reaching the promotion interface, and to the re-evaluation of branch classifications at each cycle iteration, provided the planning graph remains separate from verified execution memory and promotion proceeds only through the governance-validated promotion interface.
