≈317 tokens on demand. Your agent reads this file only when SKILL.md points to it.
Stream Collaboration Patterns
Purpose: handoff patterns between Stream and partner agents, plus common pipeline archetypes.
Contents
- Partner-agent flows
- Common orchestration patterns
Partner-Agent Flows
| Pattern |
Trigger |
Flow |
Main Output |
| Schema-to-Pipeline |
new source/target model exists |
Schema -> Stream -> Builder |
connector and load requirements |
| Analytics Pipeline |
KPIs or marts are requested |
Pulse -> Stream -> Schema |
mart and aggregation requirements |
| Pipeline Visualization |
architecture must be visualized |
Stream -> Canvas |
data-flow packet |
| Pipeline Testing |
tests are needed |
Stream -> Radar |
quality and integration test requirements |
| Cost-Aware Pipeline |
infra sizing is needed |
Stream -> Scaffold |
platform resource requirements |
Common Orchestration Patterns
| Scenario |
Pattern |
| Real-time analytics |
Source DB -> Debezium -> Kafka -> Flink -> Redis -> Dashboard |
| Data warehouse ETL |
APIs/DBs -> Airbyte -> Staging -> dbt -> Marts -> BI |
| ML feature pipeline |
Events -> Kafka -> Spark -> Feature Store -> Models |
| Event sourcing |
Commands -> Event Store -> Projections -> Read Models |
