Cloudflare is making it easier for developers to create AI agents with several new updates that were announced today.
First, the company announced a remote Model Context Protocol (MCP) server. MCP is an open standard that allows AI agents to interact with external services, which is important in enabling them to complete tasks on a user’s behalf.
According to Cloudflare, MCP has thus far been limited to running locally, which has prevented its mainstream adoption. Now, users will be able to build and deploy remote MCP servers through Cloudflare, which will allow agents to connect and interact with services without needing to rely on a locally hosted server.
RELATED: GitHub Copilot adds agent mode, MCP support in latest release
The company also has integrations with Auto0, Stytch, and WorkOS to enable users to build in authentication and authorization capabilities for delegate permissions to agents.
Another update from the company is the expansion of Durable Objects to the free tier (previously this feature was only available for paid users). Durable Objects are a type of Cloudflare Worker that combine compute and storage. According to Cloudflare, they are an important foundation for AI agents that need to maintain context across interactions.
Cloudflare also announced the general availability of Workflows, which enable users to build multi-step applications. Workflows can persistently run and retry until an action is successful, such as an agent for booking a trip that needs to be constantly searching for flights in a price range, then booking once one fits the budget.
“Cloudflare was built for this moment. First, we built the most interconnected network on the planet. Then, we built a developer platform that took advantage of that network to run code within 50 milliseconds of 95% of everyone online. And, we’re keeping our foot on the gas to give developers the best tools to build the future of agentic AI,” said Matthew Prince, co-founder and CEO of Cloudflare.
