The company says each agent has access to its own computer and can use tools, coordinate other agents, and connect to several Google apps. Users can also attach custom MCPs. These are reported product capabilities in an app preview, not independently tested performance claims.
The mechanism is a voice interface paired with agent execution and external connections: a user can assign work hands-free, then receive an update when it is ready. Sesame says its agents can help plan a trip or research a personal project. The preview is available in its iOS and Android apps, while the company’s broader glasses plan is aimed at 2027. The glasses themselves have no announced availability date. This timing matters because the preview offers a current software signal while the all-day wearable remains a future product ambition.
The first implication is that Sesame is presenting voice as an entry point for delegated, connected agent work, with the phone preview providing the evidence today. That does not establish reliability, breadth of integrations, or the eventual glasses experience. Watch whether Sesame expands app connections and developer tools, and whether users can complete useful tasks consistently across longer workflows. The next concrete signal is the promised developer tooling later this year, not an assumed glasses launch date.