Predict

Five expert personas independently analyze a proposed change and debate trade-offs before you write any code. Catches architectural, security, performance, and UX issues early.

What This Skill Does

Predict simulates a design review by running your proposal through five distinct perspectives: Architect, Security Expert, Performance Engineer, UX Specialist, and Devil’s Advocate. Each persona analyzes independently, then conflicts are identified and resolved into a clear verdict.

When to Use

  • Before implementing a major feature
  • Before risky architectural changes (database migration, auth rewrite)
  • When evaluating competing approaches
  • Before touching production-critical code paths

Personas

PersonaFocus
ArchitectSystem design, coupling, extensibility, API contracts
SecurityAttack surfaces, auth flows, data exposure, OWASP risks
PerformanceLatency, memory, N+1 queries, caching, scalability
UXUser impact, cognitive load, accessibility, error states
Devil’s AdvocateHidden assumptions, edge cases, what could go wrong

Arguments

ArgumentDescription
descriptionFeature or change proposal (natural language)
--files <glob>Optional glob to read affected code areas

Example Usage

/ck:predict "Add WebSocket support for real-time notifications"
/ck:predict "Migrate authentication from JWT to session cookies"
/ck:predict "Replace REST API with GraphQL" --files src/api/**/*.ts

Workflow

  1. Read — parse proposal and optional file paths
  2. Analyze — each persona reviews independently
  3. Agreements — identify points where 4+ personas align
  4. Conflicts — surface competing concerns between personas
  5. Resolve — weigh trade-offs for each conflict
  6. Verdict — produce final recommendation

Verdict Levels

VerdictMeaning
GOProceed with implementation — no blocking concerns
CAUTIONProceed with specific mitigations — risks are manageable
STOPDo not proceed — fundamental issues need resolution first

STOP triggers (any one forces STOP): security auth bypass detected, architect finds design incompatibility, performance analysis shows exponential scaling, or devil’s advocate surfaces a false core assumption.

Output

SectionContent
VerdictGO / CAUTION / STOP
AgreementsPoints all personas aligned on
Conflicts & ResolutionsTable with each persona’s position + recommendation
Risk SummaryRisk, Severity, Mitigation
RecommendationsNumbered action items
  • Scenario — generate edge cases to test after Predict approves
  • Plan — create implementation plan after debate
  • Cook — implement after planning