Capture the May 7 Mattermost thread updates for PDIAP-15836 / PDIAP-12284 and keep current work memory aligned with Jeff's direction.
Work Done
Mattermost sync confirmed David moved PDIAP-15836 to In Progress and found a possible minimal dismissal fix path that would not require removing UIHostingController.
Refined the Copilot implementation direction for the combined PDIAP-15836 / PDIAP-12284 branch: evaluate XFlowViewMaker-owned global host-mode resolution instead of Fid4-owned per-flow mapping.
Friday standup reported PDIAP-15836 as In Progress and aligned with PDIAP-12284 so dismissal/lifecycle changes and UIKit-removal work can be handled in the same branch and validated together.
Friday standup reported continued PDIAP-12284 evaluation around a dual-path strategy: SwiftUI host should become the default while preserving the current UIHostingController path temporarily for validation.
Findings
David's early PDIAP-15836 finding was that the tested dismissal approach appeared coupled to the UIHostingController removal path, but he was investigating whether the lifecycle/dismissal fix could be isolated.
Current working design preference is SwiftUI host by default, including missing or unknown feature configuration; UIHostingController should be selected only by an explicit temporary fallback flag.
XFlowSDK should consume a resolved host-mode decision rather than depend directly on LaunchDarkly, Flagship, or app-specific feature-flag clients.
The SwiftUI dismissal path still needs evidence that delegate/session-clear callbacks fire only after dismissal completion; do not rely on onDisappear alone unless simulator logs validate it.
Communication
Jeff directed David to do the PDIAP-15836 dismissal/lifecycle work and PDIAP-12284 UIKit-removal work in the same branch, because both changes are disruptive enough to require consumer testing.
Jeff approved the Friday standup wording for Teams, and David sent it.
Jeff said he was traveling to Managua and asked David to monitor and respond on Teams while he was gone, asking for help if a message came in that David was unsure how to answer.
Next Steps
Treat PDIAP-15836 and PDIAP-12284 as same-branch work unless later direction changes; keep consumer-testing impact explicit.
Ask Copilot to inspect whether XFlowViewMaker can use existing FeatureEnabling / Featuring abstractions for global host-mode resolution; if not, have it return the smallest dependency-injection option instead of pushing per-flow logic into Fid4.
