Back to Prompt Library
planning
Agent Roles and AutoGen Configuration
Inspect the original prompt language first, then copy or adapt it once you know how it fits your workflow.
Linked challenge: Building Adversarial Deepfake Detection
Format
Text-first
Lines
1
Sections
1
Linked challenge
Building Adversarial Deepfake Detection
Prompt source
Original prompt text with formatting preserved for inspection.
1 lines
1 sections
No variables
0 checklist items
Define the key agent roles for your deepfake detection and moderation system (e.g., Image Analyzer, Prompt Scrutinizer, Policy Enforcer, Red Team Agent). For each agent, specify its capabilities, system message, and tool access. Outline how these agents will interact within AutoGen, focusing on how 'Red Team' agents will challenge 'Moderation' agents and how consensus on moderation actions will be reached.
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.