Cloudflare enhances AI development with new MCP server

Cloudflare has unveiled advancements in AI and developer tools to facilitate the development and deployment of AI agents.

Cloudflare has introduced the industry's first remote Model Context Protocol (MCP) server, which now forms part of its Developer Platform, aiming to streamline the building and deployment of AI agents.

This system promises to remove cost and complexity barriers associated with AI agent development.

AI agents are systems enabled by AI that act autonomously and adapt to various environments, potentially offering significant productivity gains. However, these agents require a platform combining AI models, workflows, and API access. Cloudflare aims to provide these components in a scalable and cost-effective manner.

Matthew Prince, co-founder and CEO of Cloudflare, stated, "Cloudflare is the best place to build and scale AI agents. Period. The most innovative companies out there see that agents are the next big frontier in AI, and they're choosing Cloudflare because we've got everything they need to move fast and build at scale on our Workers platform. Cloudflare was built for this moment."

The introduction of the remote MCP server allows AI agents to interact directly with external services, thus enabling the completion of tasks on behalf of users.

This marks a shift from AI providing instructions to undertaking actions themselves. Previously, MCP servers operated locally, limiting broader adoption, a challenge Cloudflare seeks to overcome.

Cloudflare's collaboration with Auth0, Stytch, and WorkOS is intended to streamline the authentication and authorisation required for secure agent deployment.

Durable Objects, a mechanism that combines computation with storage, are now accessible on Cloudflare's free tier, thereby broadening access for developers.

These objects support stateful applications in a serverless environment, maintaining context across interactions. This facility could enhance the functionality of AI agents by remembering past interactions and adapting accordingly.

Cloudflare's network is equipped to handle millions of simultaneous interactions, ensuring responsive operations with low latency.

Cloudflare Workflows, which has now achieved general availability, provides developers with a reliable tool for creating and managing multi-step applications with AI integration. This includes building agent workflows for tasks such as travel booking, where various steps are automated over time.

The advancement also aims to enhance cost-efficiency in AI deployment, as Cloudflare's serverless platform scales resources in accordance with demand, thus reducing overall costs. Kate Holterhoff, Senior Analyst at RedMonk, noted, "Cloudflare offers a developer-friendly ecosystem for creating AI agents that includes a free tier for Durable Objects and serverless options for AI inference. These low-cost, easy-to-use options could empower more businesses to adopt and experiment with agentic AI."

Cloudflare has claimed a leadership stance in making AI inference accessible, with over 190 global cities equipped with GPUs to provide user-near AI services for low-latency experiences.