fix(title-proofread): never add thousand separators (Chosun style)

feat(title-proofread): add optional body context input (v2 prompt)

feat(title-proofread): expand Chosun stylebook coverage in PRIORITY 0

feat(title-proofread): never attempt to balance unbalanced quotes

feat(title-proofread): address customer feedback round 1

refactor(ui): drop v16 legacy tab

refactor: drop category selector from title proofreading tab

feat: add title proofreading sandbox tab

ui: surface v24 tab first (default), v16 legacy second

ui: remove per-tab description blurbs

fix(docker): COPY pipelines_v24.py into image

feat: add v24 pipeline (FT + self-consistency × 2 + tool-calling judge)

fix: 1-paragraph-per-bulk chunking + KSS sentence-split for overflow

fix: preserve bulk boundary \n + substring-count dedupe + 30-char prefix

fix: add paragraph_dedupe post-process to remove residual LLM echoes

fix: restore PARAGRAPH_SEP + per-bulk <원문> to suppress step1 echo

Replace 500-char limit with paragraph-greedy chunking + parallel run

Limit input to 500 chars (UI cap + handler validation)

Catch step1 LLM duplication via leading-sentence reoccurrence

Single-tab feedback UI (Good/NotBad/Critical); FT dedup tighter

Drop supabase SDK; call REST via httpx to avoid websockets conflict

Pin websockets>=13 for supabase realtime

Remove Supabase UI warnings; add startup diagnostic log

Move blindtest tab first; add RLS policies and error surfacing

initial deploy

initial commit