deBridge has announced the release of a Model Context Protocol server aimed at enabling artificial intelligence systems to initiate and complete blockchain transactions across multiple networks. The company positioned the new infrastructure as a step toward deeper automation in decentralized finance, allowing AI-driven tools to manage complex cross-chain operations without direct human involvement at each stage.
At launch, the MCP server supports 24 blockchain networks. These include major ecosystems such as Ethereum, TRON, and Base, alongside compatibility with Solana. By covering both EVM-compatible chains and non-EVM environments, deBridge aims to reduce fragmentation that has traditionally complicated automated strategies across blockchains.
Enabling AI Agents and Developer Tools
deBridge explained that the MCP server is designed to integrate directly with widely used AI and developer environments. Tools such as GitHub, Copilot, Claude, and Cursor can be connected to the protocol, enabling AI agents and developers to execute swaps, bridge assets, and coordinate multi-step transaction flows programmatically.
According to the company, this design removes the need for repeated manual actions that typically slow down cross-chain activity. Instead of switching networks, approving multiple wallet prompts, or submitting transactions one by one, execution is consolidated into a single layer. deBridge described this approach as a response to long-standing operational bottlenecks that have limited scalable automation in decentralized systems.
Deterministic Execution and MEV Awareness
The MCP server is built around deterministic execution principles. Transactions are configured in advance with predefined parameters that govern how and when settlement occurs. deBridge stated that this structure is intended to provide predictable outcomes, even when transactions span multiple chains.
Routing within the system is described as MEV-aware, meaning that execution paths are designed with consideration for miner or maximal extractable value dynamics. Pricing is quoted upfront to reflect expected execution results, giving users and AI agents clearer visibility into likely outcomes before transactions are finalized.
Non-Custodial Control With Reduced Friction
Despite the high level of automation, deBridge emphasized that users retain custody of their assets throughout the process. The protocol manages wallet coordination, retries, and cross-chain interactions within its own infrastructure, but funds are never held by the system itself. This approach is intended to balance reduced user involvement with the preservation of non-custodial control, a core principle of decentralized finance.
The company suggested that by minimizing the need for continuous user input, the MCP server allows AI systems to manage sophisticated transaction flows while still operating within user-defined constraints.
1/ Vibe Trading is here
Introducing deBridge MCP, a universal trading API for OpenClaw, Claude, Cursor, and all your favorite AI workstations pic.twitter.com/1QOboIXDno
— deBridge (@debridge) February 16, 2026
Building on Intent-Based Execution
The MCP server builds on deBridge’s earlier introduction of Bundles, an intent-based execution model released in December. Bundles allow users to specify a desired outcome, such as completing a series of cross-chain actions, while the protocol handles the underlying transaction logic. With the MCP server, this framework is extended to AI-driven systems, enabling them to trigger and manage these intents programmatically.
deBridge framed this progression as a move toward fuller chain abstraction, where execution can occur seamlessly across networks without requiring users or developers to interact with each blockchain individually.
Architecture and Ecosystem Positioning
deBridge operates using a zero-TVL, solver-driven architecture that transfers liquidity directly between networks without issuing wrapped assets. The company recently raised $5.5 million in funding from investors, including ParaFi Capital and Animoca Brands, reinforcing its position within the broader Web3 infrastructure landscape.
With the MCP server now live, deBridge is positioning its protocol as a foundation for AI-based transaction execution across a wide range of blockchains. The company indicated that embedding automation directly into cross-chain activity could play a key role in scaling decentralized systems as AI-driven strategies become more prevalent.
