solution-optimality-evaluation

FoundationJIT only

analysis-agent/task-agent/review-agent: use when feasible solutions differ in hard constraints, total-change cost, or reversibility; skip when no material option remains.

Group: Engineering workflow
Supported roles: MainAnalysisTaskReview
Source: SKILL.md

Use when

Trigger signals that make this skill the right owner

solution optimality feasible alternatives hard constraints strongest alternative total-change cost reversibility option value
algorithm complexity CPU memory IO contention throughput tail latency queue growth failure recovery maintainability evidence limits
simpler faster cheaper safer future-proof weighted score deferred optimization reopening condition

Do not use when

Anti-triggers — as binding as the triggers above

the unresolved question is whether code or structure needs to exist
architecture-wide placement is open; use architecture-tradeoff-analysis, reserving implementation-structure-design for placement inside an already selected owner
algorithm or data-structure selection is open; use algorithm-data-structure-selection
language or runtime selection is open; use language-runtime-selection
language and runtime are fixed while runtime behavior is open; use language-performance-safety
package or dependency choice is open; use package-dependency-management
The open decision is a technology-stack commitment involving a framework, platform, datastore, infrastructure component, or managed service; use technology-stack-selection
a measured bottleneck is open; use profiling

Required inputs

What must be supplied before this skill can decide anything

problem acceptance hard constraints and non-goals
selected option and strongest feasible simpler or established alternative
current workload cost ownership reversibility and comparative evidence limits

Output contract

What this skill owes the next role

solution decision with hard-constraint screen, selected option, strongest rejected alternatives, total-change cost, reversibility and option value, comparative proof and limits, reopening condition, and residual risk

Escalation signals

When this skill must hand the decision back

A feasible choice can hide a failed hard constraint recurring or exit cost irreversibility or an unsupported performance claim

Loaded by

This skill never owns a task. These professionals may pull it in as Layer 3.

high-risk-design-review

Targeted references

Each reference carries its own load condition. None of them enter context by default.

Reference	Load when	Do not load when	Required by
benchmarks-and-patterns.mdbenchmark-pattern	Feasible solution candidates differ in hard constraints total-change cost reversibility resource behavior or evidence limits	One solution is fixed by current constraints and no material alternative can change the decision	TaskReviewAnalysis