How Commerce.com and Cloudflare are getting storefronts ready for AI agents
Curated from Cloudflare Blog
Most e-commerce teams are still debating how to handle basic bot traffic, yet the reality is shifting toward autonomous agents executing complex transactions. This case study details the operational mechanics of enabling WebMCP across nearly 100,000 domains simultaneously. It is not a marketing fluff piece about "AI readiness," but a technical look at how to implement standardized interfaces without sacrificing merchant autonomy. For infrastructure engineers, the challenge of maintaining per-tenant control at that scale is where the real value lies. The architecture demonstrates how to expose tool capabilities to LLMs while preserving the security boundaries required for production environments. If you are planning to expose backend services to non-human consumers, studying this rollout strategy will save you from common pitfalls in permissioning and state management. **Takeaway:** When integrating AI agents with existing SaaS platforms, prioritize a centralized middleware layer that abstracts tenant-specific logic, ensuring you can scale capabilities without rewriting individual merchant configurations.
Almost 100,000 Commerce.com domains now give AI agents real shopping tools through WebMCP on Cloudflare. Here's how we turned it on across every store at once, while keeping each merchant in control.
— Cloudflare Blog