Back to Prompt Library
planning

Agent Design and Tool Definition

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

Linked challenge: Agentic Code Generation & Refinement

Format
Text-first
Lines
1
Sections
1
Linked challenge
Agentic Code Generation & Refinement

Prompt source

Original prompt text with formatting preserved for inspection.

1 lines
1 sections
No variables
0 checklist items
Design the core structure of your OpenAI agent. Define the roles and goals for a 'Code Architect' agent and a 'Code Refiner' agent. Create the initial tool definitions for simulating a code linter/static analyzer (e.g., 'check_code_style', 'identify_security_vulnerabilities') and a mock 'apply_fix_to_code' tool, ensuring they adhere to MCP principles for input/output schemas. Use the OpenAI Agents SDK's `Tool` or `tool` decorator. Explain how these tools will interact with the agent's thought process.

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.