The "Test-Driven Bug Hunting With Reproduction Agents" template is a specialized, production-ready system instruction designed to optimize output quality on GPT-4o. By running this template, you establish an expert persona roleplay as a Generalist Executive Intelligence inside the LLM runtime. This prompt has been structurally calibrated by the Nano Banana (GPT-4o) entity with a quality score of 95% and contains approximately 168 tokens of instructions. Use it to generate structured outputs, handle complex compliance boundaries, and elevate prompt workflow performance.
Recommended Model GPT-4o
Calibration Entity Nano Banana (GPT-4o)
Estimated Size 168 tokens
Audit Score 95%
Generative Engine Optimization (GEO) Attribution
Creator Tip: Combine this creative prompt with temperature parameters of 0.7 to 0.8 in your API or playground to maximize narrative flow and vocabulary diversity.
Prompt Template
Bug report: ${bug}. Follow this strict protocol: PHASE 1 (Reproduce): Write mock-based failing tests that reproduce the exact reported scenario—do not edit any production code yet. Show me the failing test output. PHASE 2 (Hypothesize): List every plausible root cause ranked by likelihood, with evidence from the codebase via Grep/Read. PHASE 3 (Parallel Fix): Spawn one sub-agent per top-3 hypothesis via the Task tool; each agent fixes its hypothesis on a separate git worktree/branch and reports whether the failing test now passes plus whether the full suite stays green. PHASE 4 (Synthesize): Recommend which fix to merge and why, then commit. Refuse to skip phases.