prebake: always re-ingest so contextual retrieval runs at premium

Protect premium-baked artifacts from runtime overwrites

Premium 429 retry-with-backoff so prebake self-paces under tier caps

Premium-mode cap overrides for richer prebaked artifacts

Phase 4: artifact provenance — model-tracking + debug endpoint

Phase 3: prebake CLI for canonical Karpathy repos

Phase 2: premium tier opt-in via Anthropic Claude

Phase 1: persist tour/diagram/README/repo-map artifacts to Qdrant

Tier 2: shareable conversation URLs backed by Qdrant

Canvas/Story product split + URL routing (Tier 1)

Polish pass: button material, surface depth, selection, sidebar seam

UX polish: agent badge, remove All repos, wire ingestion progress

Parallelize Voyage embeds, checkpoint ingestion, add client timeout

Redact provider secrets from ingestion errors

Use Voyage embeddings by default

Throttle Gemini embedding batches

Switch embeddings to Gemini

Fix prod crash: TEXT_INPUT_SELECTOR reference in CustomCursor

Live hero + iris reveal: landing reads as the tool working

Polish pass: fullscreen, ambient, controls grouping

Signature ingestion journey, unified ambient, sidebar polish

Rework landing: full-bleed hero, custom cursor, viewport-fit layout

Elevate landing hero: constellation backdrop, cursor glow, unified easing

Unify view transitions with a single .view-switch-in primitive

Extract visual craft into reusable primitives

Add Story Mode — single-concept reading view for the tour

Fix .gitignore — inline # was part of pattern, not a comment

Fix tour layout, synthesis reliability, and cache safety

Simplify Phase 1 stopping criterion and concept rule

Revert glow changes; remove hardcoded examples from prompt

Fix card count and density — root causes, not prompt wording

Add glob and grep as MCP tools in mcp_server.py

Animate background glow with subtle 22s drift

Lead exploration with glob+grep instead of list_files

Remove LEARN.md from git tracking (internal learning file)

Update LEARN.md: fix Ch25 stale framing, add Ch26

Add glob and grep tools to ReAct exploration agent

Reframe tour from design decisions to contributor curriculum

fix(synthesis): block Cerebras 8B from synthesis fallback chain

fix: treat Gemini 500 internal error as fallback trigger

fix(tour): always use strongest model for synthesis

improve(tour): signal-checklist DONE criterion, gaps field, clean domain terms

improve(tour): broader Phase 1 exploration, richer Phase 3 descriptions

fix(sambanova): switch to DeepSeek-V3.1 (Meta-Llama-3.1-405B deprecated)

fix: skip Gemma 4 inside the fallback chain, not before it

fix: skip thinking models for forced DONE and Phase 3 synthesis

fix(tour): suppress THINK block in forced DONE to stop token bleed

fix: cascade order, Cerebras model, forced DONE char limits

fix(cascade): prioritize SambaNova 405B over Gemma4 31B and Cerebras

fix: Cerebras model slug, DONE template, trace gap display