Mechanism
Slope-constrained speculative simulation is the rule that determines which hypothetical futures the forecasting engine is permitted to evaluate for promotion. Within the planning graph, each branch represents a distinct hypothetical trajectory: a sequence of speculative mutations, delegation outcomes, environmental transitions, or intent resolutions the agent is considering as a possible future. The slope constraint makes the agent's trust slope trajectory a structural filter over those branches. It is not a soft preference or a ranking criterion. It is a hard architectural boundary that prevents the agent from promoting any speculative branch whose execution would produce a trust slope discontinuity.
The constraint is enforced by the slope validation module, one of the five principal components of the forecasting engine. For each speculative branch the slope validation module computes a hypothetical Derived Anchor Hash, written DAH prime, representing the trust slope state that would result if the branch were promoted to execution. The branch is admitted or rejected on the basis of how that hypothetical hash relates to the agent's current trust slope trajectory, not on the basis of any statistical or learned score.
The Hypothetical Derived Anchor Hash
The slope validation module computes the hypothetical 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 mutations are never applied to verified execution memory. They are applied only to the sandboxed copy, so the computation projects what the agent's provenance would look like after the branch executed without committing anything.
The hypothetical DAH prime is then compared against the agent's current trust slope trajectory using the same continuity validation algorithm that the governance infrastructure applies to committed mutations. There is no separate, weaker check for speculative content: the forecasting engine evaluates a hypothetical future against exactly the criterion that would later judge it if it were real. If the hash chain relationship between the current DAH and the hypothetical DAH prime satisfies the cryptographic lineage requirements, continuity is maintained. If the speculative mutations would produce a lineage gap, a hash chain discontinuity, or a provenance violation, continuity is broken.
Slope-Eligible and Slope-Ineligible Branches
A branch whose hypothetical DAH prime maintains continuity with the trust slope is slope-eligible. A branch whose hypothetical DAH prime breaks continuity is slope-ineligible. Slope eligibility is a structural property of the branch, derived from the cryptographic lineage relationship, rather than a threshold on a continuous quantity.
Only slope-eligible branches may be promoted to execution. A slope-ineligible branch may still be retained in the planning graph for introspective purposes, letting the agent understand why a particular hypothetical future is structurally foreclosed, but it cannot advance through the promotion interface. This is the strict form of the constraint: regardless of how broad the agent's speculative reasoning becomes, the forecasting engine cannot produce an execution candidate that would violate the system's trust and provenance guarantees.
Beyond the binary eligibility decision, the slope projection also computes the magnitude of the trust slope continuation, how much a branch advances or retreats along the trust slope trajectory. This magnitude is used to rank slope-eligible branches comparatively by their trust slope impact, but it does not change whether a branch is eligible.
Prospective Filtering
The slope constraint operates prospectively. It filters speculative branches before they reach the promotion interface, so the governance pipeline never receives a promotion candidate that would fail trust slope validation. The promotion interface remains the sole gateway from speculative to verified status, subjecting each candidate to the full governance evaluation pipeline including policy compliance, trust slope validation, integrity impact assessment, and capability verification. The slope constraint does not replace that gateway; it ensures the gateway is only ever asked to evaluate branches that already have a viable path through it.
This prospective filtering is described as more efficient than reactive filtering, in which the agent would construct all possible branches and reject the noncompliant ones only at promotion time. By concentrating the agent's cognitive resources on branches that have a viable path to execution, prospective filtering keeps speculative work aligned with what the governance infrastructure will actually admit.
Interaction with the Integrity Field
The slope constraint interacts with the integrity field through the integrity impact projection. For each slope-eligible branch the integrity engine computes the projected integrity impact, the change to the agent's integrity score across all three domains that would result from executing the branch. Unlike slope eligibility, a negative integrity impact does not automatically disqualify a branch. Instead the magnitude of the integrity impact is incorporated into the branch's evaluation score, reducing the branch's priority relative to branches with neutral or positive integrity impact.
The two effects are therefore distinct in kind. Trust slope continuity is a hard gate that determines whether a branch may be promoted at all. Integrity impact is a graded contribution to a branch's ranking among the branches that have already passed the gate. Together they ensure the agent's speculative reasoning accounts for integrity consequences as well as governance compliance.
Branch Classification
Slope eligibility feeds the four-way branch classification the forecasting engine assigns to every branch. A branch that is slope-eligible, policy-compatible, and positively or neutrally reinforced is marked eligible: a viable candidate for promotion, ranked among its peers by a composite score that includes its trust slope continuation magnitude alongside projected outcome quality, integrity impact projection, affective reinforcement strength, and intent alignment. A branch that is slope-eligible and policy-compatible but negatively reinforced is marked introspective and retained for self-examination rather than promotion. A branch that is slope-eligible, policy-compatible, and better suited for transfer to a child agent is marked delegable. A branch that fails slope validation, fails policy compatibility, or both is marked pruned and scheduled for removal by the pruning manager.
Classification is not permanent. The trust slope trajectory changes as the agent commits mutations, and a branch's eligibility changes with it. When a committed mutation changes the agent's trust slope trajectory in a way that renders a branch's hypothetical DAH prime discontinuous, the branch is reclassified as pruned. This slope invalidation is evaluated at each forecasting execution cycle. Conversely, a branch that was slope-ineligible may be reinterpreted as slope-eligible when the agent's trust slope trajectory changes in its favor.
Determinism and Distinction from Statistical Search
The speculative mutation simulation that produces the projected outcomes is deterministic: given the same input state and mutation sequence, it produces the same projected outcome. The specification distinguishes this from statistical tree search methods such as Monte Carlo Tree Search. Rather than being evaluated statistically over random rollouts, each simulation step operates deterministically on defined structural fields, produces reproducible projected outcomes, and is constrained by trust slope continuity and policy compatibility at every step. The slope constraint is the structural element that enforces that per-step continuity discipline, which is what lets an archived planning graph support forensic reconstruction of why specific branches were foreclosed.
Disclosure Scope
Slope-constrained speculative simulation, comprising the slope validation module's computation of a hypothetical Derived Anchor Hash by applying a branch's speculative mutation sequence to a sandboxed copy of the agent's lineage, the comparison of that hypothetical hash against the agent's current trust slope trajectory using the same continuity validation algorithm applied to committed mutations, the resulting slope-eligible and slope-ineligible determination, the prospective filtering of branches before the promotion interface, the integrity impact projection, and the role of slope eligibility in the eligible, introspective, delegable, and pruned branch classification, is disclosed in the cognition filing (U.S. Application No. 19/647,395 and its international counterpart) in the description of the forecasting engine and its slope-constrained speculative simulation. This article describes that disclosed mechanism and does not constitute a complete enabling disclosure. The scope extends to embodiments in which the trust slope continuity criterion is realized over different lineage and hash representations, provided the constraint remains a hard structural filter that admits only branches whose projected provenance maintains continuity with the agent's current trust slope trajectory.
