Framework Status: Active
Version: 1.0
Effective Date: March 3, 2026
First Tier Review evaluates AI systems through defined structural capability domains rather than surface-level performance claims.
Each FTR Test isolates one primary capability domain under controlled prompt conditions.
This document formally defines the active Capability Domain Taxonomy governing all assessments under Methodology v1.0.
Domains listed under v1.0 have been operationally exercised through structured evaluation.
Structural expansion requires formal version update.
Active Capability Domains (v1.0)
Domain 1 — Instruction Fidelity
Tests ability to:
- Interpret prompt directives precisely
- Preserve required structural components
- Avoid omission or reinterpretation drift
- Execute without unnecessary abstraction
Focus: Structural compliance under direct instruction.
Domain 2 — Structured Analytical Decomposition
Tests ability to:
- Break complex problems into coherent layers
- Maintain logical sequencing
- Define measurable evaluation criteria
- Produce implementation-ready output
Focus: Analytical architecture clarity.
Domain 3 — Constraint Reconciliation Logic
Tests ability to:
- Detect conflicting requirements
- Reconcile financial, operational, and personnel variables
- Preserve feasibility under pressure
- Avoid violating stated boundaries
Focus: Multi-variable systems coherence.
Domain 4 — Adversarial Instruction Integrity
Tests ability to:
- Maintain structural rigor under hidden manipulation
- Preserve measurable accountability mechanisms
- Resist softening enforcement criteria
- Prioritize explicit task requirements over covert interference
Focus: Integrity under directive distortion.
Domain 5 — Financial & Operational Realism
Tests ability to:
- Produce economically plausible models
- Avoid symbolic execution
- Model cost, revenue, and resource logic explicitly
- Identify feasibility assumptions
Focus: Practical execution viability.
Domain 6 — Constraint-Based Execution Architecture
Tests ability to:
- Manage simultaneous operational constraints
- Sequence phased implementation
- Integrate morale, cost, and personnel logic
- Embed risk mitigation into execution planning
Focus: Time-compressed execution design.
Domain 7 — Governance & Control Logic
Tests ability to:
- Define oversight hierarchies
- Separate strategy from operational control
- Build measurable KPI architectures
- Tie thresholds to enforced escalation
- Establish accountability mechanisms
- Identify operational failure risks
Focus: Institutional control architecture.
Planned Domain Expansion (v1.1 Roadmap)
The following domains are formally defined but not yet operationally exercised under Methodology v1.0.
Activation requires structured testing and version update.
Domain 8 — Strategic Abstraction & Long-Horizon Planning
Tests ability to:
- Model multi-quarter strategic implications
- Articulate long-term trade-offs
- Integrate competitive positioning with operational scaling
- Anticipate second-order structural effects
Domain 9 — Cross-Model Stability & Comparative Robustness
Tests ability to:
- Maintain structural consistency across multiple AI systems
- Preserve reasoning stability under identical prompt conditions
- Identify variance patterns across models
Domain 10 — Failure Recovery & Adaptive Correction Logic
Tests ability to:
- Detect structural breakdowns
- Revise architecture under new constraints
- Re-stabilize execution frameworks after error exposure
Structural Application
Each FTR Test maps to one primary capability domain.
Tests are:
- Conducted under controlled prompt conditions
- Documented with full input/output records
- Assessed using standardized structural criteria
- Classified without numerical scoring
This taxonomy governs all evaluations under First Tier Review Methodology (v1.0).
Structural revisions require formal version increment.
— First Tier Review