Follow up to: #3582494 _**NOTE: Before working on this issue, verify it's still relevant and makes sense.**_ **Description** This discussion captures future work that could help administrators understand, debug, and improve context selection quality. Potential areas to discuss: - Context selection preview/testing UI - Token count display for context items - Plugin-based or event-based selection pipeline - Multi-scorer ensemble selection - Conflict detection and resolution - Context quality dashboard - Selection diagnostics and inclusion/exclusion reasoning Questions to resolve later: - What selection debugging information should be visible to administrators? - Should preview, quality scoring, and diagnostics share one data model? - How should token counts be estimated and displayed? - Which quality signals should affect actual selection versus only admin reporting? This should remain postponed until the selection extension points and diagnostics discussion is resolved. [x] AI Assisted Issue