Una plataforma SaaS que integre flujos de aprobación humana en tiempo real dentro de ChatOps para acciones de agentes automatizados.

Repository: archestra-ai/archestra Author: joeyorlando # Background Archestra already supports policies that can require approval for certain tool calls, but autonomous sessions like A2A, Slack, MS Teams, and sub-agents currently do not have an approval path. In practice that means approval-required actions get blocked instead of paused and reviewed. ChatOps feels like the obvious place to close that gap. A user should be able to get an approval request from a ChatOps-triggered flow and then approve or decline it in a safe Archestra surface. ## Desired Outcome Let approval-required agent actions pause and resume cleanly when they come from ChatOps. ## Implementation Notes This will require some research around approval UX, secure deep links, actor verification, expiration/retry behavior, and how pending approvals are represented across chat, ChatOps, and policy enforcement. The exact design should be up to the implementer based on the best approach. --- <!-- archestra-banner:v1 --> <a href="https://archestra.ai/contributor-onboard" rel="nofollow noreferrer noopener" target="_blank"> <img width="1954" height="227" alt="Image" src="https://github.com/user-attachments/assets/1f9497f2-96ad-4334-b3a4-c63ff38abdb9"/> </a>