Back to Prompt Library
planning
Initial Agent System Design
Inspect the original prompt language first, then copy or adapt it once you know how it fits your workflow.
Linked challenge: AI Code Audit & Optimization Agent
Format
Text-first
Lines
1
Sections
1
Linked challenge
AI Code Audit & Optimization Agent
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 for code auditing using the OpenAI Agents SDK. Define at least three distinct agent roles (e.g., Code Analyst, Security Reviewer, Performance Optimizer) and their primary responsibilities. Describe how these agents will communicate and pass information, focusing on the Assistants API's `run_steps` and `tool_outputs` concepts. Specify the tools each agent might need, referencing `openai.beta.assistants.ToolResources`.
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.