This study introduces a multi-agent reinforcement learning approach using the MegaAgent framework to optimize decision-making in oil and gas extraction. Traditional methods struggle with optimizing parameters like well pressure and extraction rates. By employing the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm, this research enhances efficiency through intelligent optimization of pressure and extraction controls. The framework supports efficient agent collaboration via task decomposition and parallel execution, improving overall system performance. Validation in a simulated environment shows that the proposed method surpasses traditional single-agent reinforcement learning techniques by better coordinating multiple agents, thus boosting oil extraction effectiveness and reducing equipment wear.