deBridge launches MCP, enabling AI agents to execute non-custodial cross-chain transactions
Cross-chain protocol deBridge (DBR) has launched a Model Context Protocol server, which it says is designed to let AI agents and developer tools like Claude, Cursor, and Copilot execute swaps, bridging, and multi-step onchain flows across EVM-compatible chains and Solana.
The MCP server provides what deBridge describes as deterministic execution, with quotes designed to execute as intended and MEV-aware routing optimized for reliability, according to a statement shared with The Block on Monday. Users retain custody of their funds throughout the process, with execution abstracted behind a single interface that handles wallet orchestration, chain switching, and transaction retries — functions that the protocol said traditionally create friction for automated systems.
According to the statement, potential applications include AI trading assistants that rebalance portfolios across chains, bots executing multi-step onchain strategies, consumer apps embedding cross-chain execution, and developer tools that translate natural language into onchain actions.
MCP follows the launch of deBridge Bundles in December, an intent-based execution model that allows users to specify desired outcomes, which the protocol then carries out without direct interaction with the underlying blockchains. Bundles formed the first stage of deBridge’s broader chain-abstracted execution roadmap, which the project describes as “any action, any chain, one execution engine.”
Founded in 2022, deBridge operates a zero-TVL, solver-driven architecture that facilitates direct liquidity transfers between chains without the use of wrapped assets. The protocol currently supports 24 blockchains, including Ethereum, Base, and Tron. To date, deBridge has raised $5.5 million in funding from investors, including Animoca Brands and ParaFi Capital, as The Block previously reported.
Disclaimer: The content of this article solely reflects the author's opinion and does not represent the platform in any capacity. This article is not intended to serve as a reference for making investment decisions.
You may also like
Potential Governance Risks Behind Productivity! Goldman Sachs (GS.US) Restricts Non-Technical Employees from Using Advanced AI Programming Tools
According to reports, Goldman Sachs has banned some employees from using advanced artificial intelligence (AI) tools designed for programming after assessing concerns regarding productivity and risk.
Two Completely Opposite Bear Market Narratives Coexist in the Market: "Insufficient Hashrate vs. Excessive Hashrate"
Goldman Sachs highlights a dilemma in the AI market: while logic about power shortages and logic about excess computing power seem mutually exclusive, both have become reasons for AI trade sell-offs. Although demand from cloud providers remains strong, capital expenditures are projected to reach $1.2 trillion, with negative cash flow compounded by high leverage. A short squeeze rebound may occur at year-end, but the rebound could also present a selling opportunity.
Q3 results exceed expectations + full-year guidance raised! Grab (GRAB.US) receives bullish rating from Citigroup after earnings: Stock price can still rise over 92%
Citi has given Grab a "Buy" rating with a target price of $5.50. This target represents a potential upside of over 92% compared to Monday's closing price of $2.86.
Crypto prices
More
