Back to Prompt Library
planning

Agent System Design

Inspect the original prompt language first, then copy or adapt it once you know how it fits your workflow.

Linked challenge: FinReg Sentinel: AI Agent for Crypto Compliance

Format
Text-first
Lines
1
Sections
1
Linked challenge
FinReg Sentinel: AI Agent for Crypto Compliance

Prompt source

Original prompt text with formatting preserved for inspection.

1 lines
1 sections
No variables
0 checklist items
Design a multi-agent system architecture using Vellum Canvas for a 'FinReg Sentinel' that performs regulatory monitoring. Define the roles of each agent (e.g., 'Document Analyst', 'Risk Assessor', 'Report Generator'), their communication protocols, and the overall workflow including steps for data ingestion, processing, analysis, and output generation. Specify how Durable will be integrated for persistent context.

Adaptation plan

Keep the source stable, then change the prompt in a predictable order so the next run is easier to evaluate.

Keep stable

Preserve the role framing, objective, and reporting structure so comparison runs stay coherent.

Tune next

Swap in your own domain constraints, anomaly thresholds, and examples before you branch variants.

Verify after

Check whether the prompt asks for the right evidence, confidence signal, and escalation path.