Background: Blockchain and Multi-Agent System (MAS) are increasingly combined to support decentralized, secure, and autonomous peer-to-peer energy trading in microgrid environments. Objectives: This systematic review investigates how blockchain and MAS are integrated to support microgrid energy trading, identifies architectural and operational models, examines real-world implementations, and highlights technical, regulatory, and security challenges. Unlike prior reviews that focus on blockchain or MAS in isolation, this study provides a unified and comparative analysis of their joint integration. Methods: Following PRISMA 2020 guidelines, a systematic search was conducted in IEEE Xplore, ACM Digital Library, and ScienceDirect, with the last search performed on 10 January 2025. Eligible studies focused on blockchain–MAS integration in microgrid energy trading; non-energy and non-microgrid applications were excluded. Study selection was performed independently by two reviewers, and methodological quality was assessed using an adapted Joanna Briggs Institute (JBI) checklist. A narrative synthesis categorized integration levels, blockchain platforms, MAS roles, and implementation contexts. Results: A total of 104 studies were included. Three dominant integration levels were identified—basic, intermediate, and advanced—distinguished by how decision-making responsibilities are distributed between MAS and smart contracts. Ethereum and Hyperledger Fabric were the most commonly used platforms. MAS agents perform concrete operational functions such as bid and offer generation, price negotiation, matching, and local energy optimization, fundamentally transforming control and monitoring processes. By enabling distributed, intelligent agents to perform real-time sensing, analysis, and response, an MAS enhances system resilience and adaptability. This architecture allows for proactive fault detection, dynamic resource allocation, and coherent, large-scale operations without centralized bottlenecks. Blockchain ensured transparency, trust, and secure transaction execution. Major challenges include scalability constraints, interoperability limitations with legacy grids, regulatory uncertainty, and real-time performance issues. Limitations: Most included studies were simulation-based, with limited real-world deployment and substantial heterogeneity in evaluation metrics. Conclusions: Blockchain–MAS integration shows strong potential for secure, transparent, and decentralized microgrid energy trading. Addressing scalability, regulatory frameworks, and interoperability is essential for large-scale adoption. Future research should emphasize real-world validation, standardized integration architectures, and AI-enabled MAS optimization. Funding: No external funding. Registration: This systematic review was not registered.