Platform Capabilities

Cascade Platform provides 35+ enterprise-grade capabilities for building scalable, secure applications.

Capability Summary

All MVP capabilities are production-ready and battle-tested in enterprise deployments.

Quick Selection Guide

What are you building?

• → Long-running workflows - Process Orchestration
• → Dynamic forms & dashboards - UI Rendering
• → Business rules engine - Policy Evaluation
• → AI-powered features - Agent Integration
• → Event-driven system - Unified Event System
• → Multi-tenant app - Multi-Tenancy & Isolation

Tier 0: Core Orchestration (8 Capabilities)

Foundation capabilities for workflow execution and business logic.

Minimum capabilities needed: 1, 3, 8

Developer Experience (8 Capabilities)

Tools and infrastructure for productive development.

For local development: 12, 14, 13

Security & Governance (6 Capabilities)

Permission, compliance, and audit capabilities.

Enterprise requirements: 17, 19, 21, 22

UI & Extensibility (8 Capabilities)

User interface rendering and AI integration.

For user workflows: 23, 24 For AI features: 26, 28

Resilience & Operations (7 Capabilities)

Performance, reliability, and operational observability.

For production: 31, 32, 33, 35

Detailed Capability Matrix

Core Orchestration

Process Orchestration (MVP, Production-Ready)

What: Durable workflow execution with state persistence
When: Long-running business processes
Tech: Temporal + PostgreSQL
Guarantees: 99.5% SLO, automatic retries, event sourcing
Performance: P95 < 10s, 1000+ workflows/sec
Tests: 5
Learn More →

Policy Evaluation Framework (MVP, Production-Ready)

What: Multi-engine business rules evaluation
When: Complex business logic, compliance checks
Engines: OPA (<5ms), DMN (10-50ms), WASM (future, <1ms)
Architecture: 4-tier decision strategy
Tests: 12
Learn More →

Unified Event System (Post-MVP, Planned)

What: Event-driven workflow triggering and inter-process communication
When: Reactive workflows, webhook integrations
Technology: NATS JetStream (planned)
Features: Multi-event waiting, event schema registry
Status: Planned for Q1 2026
Tests: 0 (not yet implemented)

CDL Template Engine (Post-MVP, Beta)

What: 2-layer workflow architecture for business user configuration
When: Business users need to customize workflows
Layers: Developer templates + business user configurations
Technology: JSON Schema + CUE validation
Status: Beta, targeted for Q1 2026
Tests: 0 (in development)

UI & Extensibility

Universal Schema-Driven UI Rendering (MVP, Production-Ready)

What: Render forms, tables, charts, layouts from JSON Schema
When: Dynamic user interfaces, configuration UIs
Adapters:

• Appsmith (primary): forms, tables, charts, layouts
• RJSF (secondary): forms only, lightweight
• ECharts: data visualization
• TanStack Table: data tables

Performance: 42ms cache hit, 127ms cache miss
Tests: 45 (31 unit + 9 integration + 5 contract)
Learn More →

AI Agent Integration (MVP, Beta)

What: LLM agents integrated as workflow activities
When: AI-powered decision making, sentiment analysis
Architecture: Agents as activities (maintains determinism)
Providers: LangChain, CrewAI (planned)
Memory: v5.1 includes KV, conversation, semantic
Tests: 4
Learn More →

Developer Experience

Comprehensive CLI (MVP, Production-Ready)

What: Command-line tool for development, debugging, operations
Commands:

• App management: cascade app apply|delete|list|inspect
• Process management: cascade process start|signal|cancel
• Monitoring: cascade logs|trace|metrics
• Configuration: cascade config get|set

Performance: Sub-second for most operations
Tests: 14
Learn More →

Local Development Environment (MVP, Production-Ready)

What: One-command local infrastructure setup
Tools: Mage + Tilt
Features: Sub-3s hot-reload, automatic dependency management
Includes: PostgreSQL, Redis, Temporal, NATS, Ory, SpiceDB, Grafana
Tests: 2
Learn More →

Security & Governance

Unified Authorization Framework (MVP, Production-Ready)

What: Fine-grained, relationship-based access control
Technology: SpiceDB (Zanzibar model)
Features: Delegated access, resource ownership, relationship caching
Supports: Role-based, attribute-based, relationship-based
Tests: 3
Learn More →

Data Isolation & Multi-Tenancy (MVP, Production-Ready)

What: Complete tenant isolation at database level
Technology: PostgreSQL Row-Level Security (RLS)
Features: Transparent isolation, automatic query filtering
Guarantees: No cross-tenant data leaks
Tests: 4
Learn More →

Resilience & Operations

Comprehensive Observability (MVP, Production-Ready)

What: End-to-end monitoring via OpenTelemetry
Signals: Metrics (Prometheus), Logs (Loki), Traces (Jaeger)
Coverage: All components instrumented
Features: Distributed tracing, performance profiling
Tests: 5
Learn More →

Validation Strategy (MVP, Production-Ready)

What: Hybrid validation approach for different scenarios
Layers:

1. CUE - CDL artifact validation, schema evolution
2. JSON Schema - Business user input, UI generation
3. OPA - Business rules, context-aware policies
4. Go - Programmatic logic, type safety

Tests: 10
Learn More →

Maturity Levels Explained

MVP (Minimum Viable Product)

• ✅ Feature-complete implementation
• ✅ 80%+ test coverage
• ✅ Production deployments active
• ✅ Stable API (v1+)
• Recommendation: Use for critical systems

Post-MVP

• ⚠️ Core features complete, enhancements planned
• ⚠️ 50%+ test coverage
• ⚠️ Beta deployments
• ⚠️ API may change slightly
• Recommendation: Use for non-critical features

Enhancement

• 🔄 Platform feature, not first-priority
• 🔄 Design phase or partial implementation
• Recommendation: Plan but don’t require immediately

Future

• 🔮 Under research, not scheduled
• 🔮 <5% implementation
• Recommendation: Monitor for future versions

Capability Dependencies

Fundamental (required for all):

• Process Orchestration
• Application Lifecycle
• Secret Management

For Enterprise Production:

• All security capabilities (17, 19, 21, 22)
• Comprehensive Observability
• Disaster Recovery
• Audit Logging

For Modern Development:

• Local Development Environment
• Comprehensive CLI
• Developer Debugging
• Distributed Tracing

For AI Applications:

• Process Orchestration
• AI Agent Integration
• Agent Memory Capabilities
• Context Management

For Advanced Workflows:

• Policy Evaluation
• Unified Event System (when available)
• WASM Runtime (when available)

Performance & Scale

Test Coverage

Overall: 78.3% test coverage (347 tests)

Capability Selection Matrix

                        Simple              Complex
                        ├────────────────────┤
High Throughput  →  Choice States     →  Parallel Tasks
                →  OPA Policies      →  WASM Policies
                →  UI Rendering      →  Custom Charts

Long Duration    →  HumanTasks        →  Agent Workflows
                →  Wait States       →  Event System
                →  Temporal          →  Complex Sagas

Compliance       →  Audit Logging     →  Multi-Tenancy
Required         →  Authorization     →  Data Isolation
                →  Encryption        →  RLS + Policies

Deployment Readiness

Next Steps

Just getting started? → Getting Started Guide

Want to explore specific capabilities? → Top 5 Capabilities in Detail

Ready to build? → First Workflow Tutorial

Need reference? → CSL Syntax Reference

Support Matrix

Updated: October 29, 2025
Version: 3.1
Total Capabilities: 35
Production-Ready: 21
Beta/Planned: 14