feat(ai): Tutor Agent web UI — Smart session with visible reasoning (AI-Agent-2)#398
Merged
Conversation
…AI-Agent-2) Frontend for the Tutor agent: "Smart session" on the Vocabulary page surfaces the tutor's plan + per-item why + reading nudge, studies the planned cards via the existing FlashCard, and re-plans on HITL feedback until done -> summary. New TutorSessionPage (/:lang/vocabulary/tutor), useTutorSession state machine, TutorPlanView, tutor.ts client. Backend (anti-join fix): the UI no longer re-fetches+joins vocab (which dropped planned cards for >100-word users via the 100-cap + bad sort). TutorEndpoints loads each plan item's card scoped to the caller (Id IN ids AND UserId==userId — a 2nd anti-hallucination/isolation re-check) and enriches TutorPlanItemDto with translation/definition/sentence/bookTitle/hint/distractors. Re-plan turn cap (MaxTurns=6 server + 8-round client backstop) prevents infinite loops. UI hardening (adversarial QA): AbortController/mounted-guard, feedback retry re-submits the SAME session (no progress loss), LLM strings line-clamped + unknown-exerciseType fallback. tsc clean; 584 web + 35 backend Tutor + AiEvals green; vite build green; browser-checked (entry->plan->study->re-plan->summary + edge states, 0 console errors). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The web UI for the Tutor agent (frontend slice; backend shipped in #397). Adds backend DTO enrichment so cards render straight from the plan. architect → frontend → adversarial QA → fix cycle.
What
TutorSessionPage(/:lang/vocabulary/tutor).rationale, each item's word + exercise-type badge + difficulty + per-itemwhy, and the closingreadingNudge. Visible reasoning, not debug text.FlashCard(word → flip → Got it / Missed it;responseTimeMsmeasured)./feedbackre-plans the remainder ("your tutor adjusted your plan") until an empty plan → summary (studied / accuracy / nudge).useTutorSession(planning→plan→study→…→summary+ empty / error / signIn).Backend fix (the blocker QA caught)
The UI used to re-fetch the user's vocab and join by id to render cards — which silently dropped planned cards for users with >100 words (server caps
getWordsat 100 + a non-existent'recent'sort that ordered by creation date, not SRS due-ness). Fixed at the source:TutorEndpointsnow loads each plan item's card scoped to the caller (Id IN ids AND UserId == userId— a second anti-hallucination/isolation re-check) and enrichesTutorPlanItemDtowithtranslation/definition/sentence/bookTitle/hint/distractors. The client renders directly from the plan — no join, nothing dropped. Plus a re-plan turn cap (MaxTurns=6server + 8-round client backstop) so a persistently-missed card can't loop forever.UI hardening (adversarial QA)
AbortController/mounted-guard (no setState-after-unmount); feedback-failure retry re-submits the same session (doesn't lose progress); untrusted LLM strings (why/rationale/nudge) line-clamped with an unknown-exerciseTypefallback (no raw i18n-key leak).Verify
tscclean · 584 web + 35 backend Tutor + AiEvals green ·vite buildgreen · browser-checked (entry→plan→study→re-plan→summary, + empty / unknown-type / long-text / unmount-mid-plan, 0 console errors). No migration.Deferred
Mobile Tutor UI, SSE plan streaming, generated MC exercises beyond the existing card, admin replay link.
🤖 Generated with Claude Code