outages.live turns cloud status pages into a single live feed for your agents

Keeping track of outages across all the services a modern app depends on is hard enough for humans, let alone for AI agents that are supposed to reason about failures in real time. outages.live is a new developer‑focused service that aggregates incident data from the official status pages of hundreds of providers into a single live feed your agents can query.

The service continuously polls cloud, payments, AI, and developer platforms—including names like AWS, Cloudflare, Stripe, Twilio, GitHub, OpenAI, and Anthropic—and normalizes their incident updates. You can browse a web dashboard showing which providers are currently degraded, but the key feature is that outages.live also exposes itself as a remote Model Context Protocol server. By adding it to your configuration under the name “outages,” any compatible AI agent can ask programmatically which providers are impacted.

That makes it easier for chatbots and ops automations to distinguish between bugs in your own code and upstream outages, and to generate user‑facing explanations grounded in current incident reports. As teams experiment with letting AI handle more of their incident response, outages.live is emerging as a kind of shared outage dashboard for both human SREs and agents, and is drawing attention in developer circles abroad.