[codex] Flatten Library evidence index

[jon-devlapaz]

Context & Intent

• What problem does this solve? Library rendered learner-owned concepts as promotional cards with source and generated-map metadata, weakening the evidence-index hierarchy.
• Which agent or track does this stem from (e.g., Theta research, MVP stabilization)? Production UX audit and supervised Library refinement lane.

Implementation Details

• Brief overview of technical changes (files touched, key logic). Render one divided row per concept, keep only the derived state and strongest learner reconstruction, remove unused source/count/CTA markup, and add focused mobile browser coverage.
• Note any specific architectural boundaries crossed (e.g., frontend graph logic vs. backend drill routing). Frontend Library rendering and styling only; no training writes, SEDA projection, persistence, or backend routing changed.

MVP / Deployment Risk Check

• Does this logic rely on local behavior that might fail in Vercel serverless? No; it is static frontend rendering over the existing browser-local training store.
• Are there SSRF or error-leakage risks in new external calls? No new external calls.
• If dealing with ingestion (e.g., YouTube), is the manual fallback preserved? Not applicable; ingestion is unchanged.

UX Framework Alignment (For UI/Drill changes)

• Does this strictly enforce "Generation Before Recognition"? Yes; Library shows learner reconstruction and never reveals generated study content as evidence.
• Does the graph still tell the truth (no faking mastery)? Yes; visible state remains derived from existing training evidence.
• Is the active cognitive target explicitly clear to the user? Yes; each row exposes one concept, its earned state, and the learner's own reconstruction.

Branch: feat/library-minimal-surface
Base: main
Diff summary: 8 files changed, 176 insertions, 63 deletions; flat Library evidence rows, renderer cleanup, cache pins, and focused tests.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
learn-ops-tamagachi	Ready	Preview, Comment	Jul 13, 2026 9:11pm