All skills
performance-budgeting
FoundationJIT onlyanalysis-agent/task-agent/review-agent: use when latency, throughput, bundle, memory, CPU, query, rendering, or resource cost needs a budget; skip without performance risk.
- Group
- Reliability & operations
- Supported roles
- MainAnalysisTaskReview
- Source
- SKILL.md
Use when
Trigger signals that make this skill the right owner
- set latency throughput memory cloud cost query scan per feature cost and rendering performance budgets
Do not use when
Anti-triggers — as binding as the triggers above
- no task-local performance budgeting decision is required
Required inputs
What must be supplied before this skill can decide anything
- current task contract
- selected primary Professional Skill
- task-local trigger evidence
Output contract
What this skill owes the next role
- authorized performance and unit-cost budget with representative workload, changed-surface metrics, measurement-integrity evidence, risk-based enforcement, correctness guard, and explicit removable exception path
Escalation signals
When this skill must hand the decision back
- No budget makes performance or cost regressions subjective and late to detect
Loaded by
This skill never owns a task. These professionals may pull it in as Layer 3.
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 | Competing threshold workload capacity or enforcement patterns remain viable. | Current authority and representative measurements determine one budget. | TaskReviewAnalysis |
| checklist.mddecision-checklist | Several changed surfaces scale drivers validators or exception decisions must close together. | One measured surface has an authorized threshold validator and owner. | TaskReviewAnalysis |
| evidence-patterns.mdevidence-pattern | Threshold authority workload representativeness changed-surface coverage measurement integrity correctness exception removal or final-edit freshness claims need proof. | Fresh evidence proves authority workload coverage measurement integrity correctness and exception removal for the changed surface. | TaskReviewAnalysis |