Mind · Strategy
RootSolver
Apex triage node of a SolverTree, with Pathway Memory as its compounding learning site
Definition
Mechanism
The apex node of a solver Tree — the unique entry point where a Problem (attached to a Task) enters that tree. Its cognitive operation is triage: determining what kind of problem this is and routing it via a dispatch function conditioned on its PathwayMemory. The RootSolver holds four unique authorities: Problem Framing (initial constraints and ProblemSpace), Budget Allocation (cascading Budget to child SolverNodes), Ultimate Accountability (owning the final Result or Solution), and Pathway Memory maintenance (compounding learning about which routes work for which problem types). Every tree has exactly one RootSolver; sub-trees carry their own. A FrameError arriving from any descendant is the signal that reframing is required rather than retry; the RootSolver is its terminal consumer, since it alone holds Reframe Authority.
Invariants
- Singleton per tree: every SolverTree has exactly one RootSolver.
- Owns the final success or failure of the root Task.
- Sole reframe authority, and terminal consumer of any FrameError raised beneath it.
- PathwayMemory writes name the node that wrote them.
Risks
Failure modes
Bad Frame
problem framed incorrectly, unsolvable by downstream nodes.
Reframe Failure
Root fails to find a valid alternative frame after downstream failure.
Budget Misallocation
resources distributed poorly across child nodes.
Poisoned Pathway Memory silently biases routing decisions — mitigated by signed writes only.
Relationships
Builds on
Composes with
Learned routing memory: problem_class → route → outcome_quality
References
Infrastructure
Resource allocation limit
Typed failure signaling that lateral reframing, not retry, is required
Infrastructure
A state gap with a cost of inaction
Infrastructure
Bounded domain of a problem
Infrastructure
The output of agentic transformation
Infrastructure
The fractal unit of completed work
Society
Stateful container for a running solver instance
Infrastructure
The atomic unit of intent
Infrastructure
Branching reasoning topology
Used by
References
Approach any problem through six dimensions: scope, evidence, constraints, stakeholders, mechanism and dynamics
Before decomposing a concrete problem, locate it by climbing specialization links to the root, then retrace the route
The active strategist that frames the problem
Society
Active hierarchy of coordinated solver instances
Reference
Data schema
{
"type": "object",
"required": [
"task_ref",
"problem_frame"
],
"properties": {
"task_ref": {
"type": "string",
"description": "Reference to originating task"
},
"problem_frame": {
"type": "object",
"description": "Initial problem constraints and space"
},
"budget_allocation": {
"type": "object",
"description": "Resource distribution to children"
},
"pathway_memory_ref": {
"type": "string",
"description": "Reference to this Root's PathwayMemory"
},
"children": {
"type": "array",
"items": {
"type": "string"
},
"description": "Child solver node refs"
}
}
}