Cloudflare Introduces Web Search API via AI Gateway
Cloudflare is expanding its AI Gateway capabilities by introducing native web search API integration, allowing developers to inject real-time web context directly into model inference calls.

Bringing Real-Time Context to AI Models
When autonomous agents need to fetch live web pages, they frequently resort to guessing URLs and executing tool calls to curl them, often resulting in 404 Not Found errors when the guess is incorrect. To solve this inefficiency, Cloudflare has announced native web search API integration via AI Gateway in partnership with Ceramic.ai, Exa, and Linkup. More details can be found on the web search developer docs.
Because AI models are typically trained and frozen at a specific point in time, they operate strictly on information existing prior to their knowledge cutoff date. This restriction makes it difficult for models to interact naturally with recent events, changing application programming interfaces, or fast-evolving news developments. By integrating web search directly into the inference pipeline, applications receive fresh, structured snippets from the web injected straight into the model's context layer.

Setting Standards for Responsible Web Crawling
Cloudflare emphasizes that web crawlers should remain transparent, honest, and respectful of all bot rules and preferences, while giving site owners meaningful control over their content. Under the new launch, search partners have committed to meeting strict bot crawling standards. The crawler utilized by the web search provider must comply with Cloudflare's publicly stated requirements for verified bots, as defined in our developer documentation, and responses must provide a link pointing back to the crawled content location.
These established rules aim to foster a fair internet environment, allowing content creators to decide how their data is handled. By utilizing these web search integrations, consumers draw search knowledge from operators committed to transparency, control, and visibility by identifying crawlers, respecting robots.txt files, and citing original sources.
Integration and Control via AI Gateway
Cloudflare positions AI Gateway as the flagship integration point and control plane for the new Web Search API product. The platform offers built-in observability, unified billing, security, and access controls. Users can consume web search capabilities using their existing AI Gateway credits, collect logs and request data on search calls, and configure access permissions for different web search providers.
Furthermore, all requests surface in standard AI Gateway observability logs, and search queries draw directly from the active credit balance. Cloudflare identifies partners supporting Zero Data Retention to ensure data privacy, offering web search directly at list API pricing from partners without adding any extra markup. Additional configuration guidance is available through the developer docs.

Flexible Deployment Options for Developers
The platform supports Bring-Your-Own-Key integration with web search providers, mirroring how model inference providers are handled. This allows organizations to leverage existing setups within AI Gateway seamlessly. Developers making HTTP calls from mobile applications, existing backends, or external services can query the web search feature directly through a standard REST endpoint by passing an AI Gateway authentication token and specifying the preferred provider.
For developers building directly on Cloudflare Workers, integration requires minimal code through a standalone worker binding. Cloudflare is also actively developing native server tools directly into AI Gateway to remove the need for manual tool definitions, with web search slated as one of the earliest incorporated features. Users can begin experimenting immediately with their AI Gateway account and key on the AI Playground.
Sources
- Cloudflare BlogIntroducing Web Search API via AI Gateway
Continue chronologically




