exeris-bench-task-classifier

name	exeris-bench-task-classifier
description	Classify Exeris benchmark requests by family, target, and comparison axis. Use for benchmark triage before architecture, implementation, or reporting work.
argument-hint	Describe benchmark goal, target (Community/Enterprise/Spring), protocol mode, and expected comparison.

Exeris Bench Task Classifier

Use When

• A new benchmark request arrives and scope is unclear.
• A task mixes runtime/micro/docs concerns.
• You need normalized routing metadata.

Output Schema

• benchmark_family: MICRO | RUNTIME | COMPAT | RESULTS | DOCS_REPORTING | MULTI_DOMAIN
• target_scope: KERNEL_COMMUNITY | KERNEL_ENTERPRISE | SPRING_RUNTIME | MULTI_TARGET
• comparison_axis: WITHIN_TIER_PROTOCOL | CROSS_TIER_SAME_PROTOCOL | MODE_COMPARISON | HISTORICAL_REGRESSION | NONE
• risk_flags: list of detected primary risks

Procedure

1. Extract explicit benchmark objective.
2. Map task to exactly one primary family (or MULTI_DOMAIN).
3. Map target scope from scenario/runtime context.
4. Determine comparison axis from expected claim.
5. Emit risk flags for fairness, reproducibility, or confidentiality.