Open-source AI agent team platform Select launches, supporting dynamic autonomy management based on delivery credentials

On November 7, 2026 at 09:53, a developer posted on DEV Community's AI section to announce the open-source launch of Select, a platform for managing AI agent teams. The author framed the motivation this way: AI agents keep getting more capable, yet it remains hard to tell whether a given agent actually deserves to work autonomously. Most orchestration platforms, the author noted, take the conventional route by letting users configure permissions and decide up front how autonomous their agents should be, and Select was built to explore a different approach instead. According to the post, Select is constructed with Next.js, FastAPI, PostgreSQL and Docker. Rather than granting agents autonomy from the outset, it performs delivery-contract verification on completed tasks through deterministic rules, then adjusts permissions dynamically based on the verification evidence produced. Only one report on Select is available so far, and there is no additional information about the project's subsequent development.