Cap Telegram flood-wait sleep so long RetryAfter cannot hold the chat lock

Review fixes: full sender names, nick hints, ctx persist, trim order

Group context: author tags and reply-target answers

Chat memory: sender directory and combined 100-message cap

Fix small hardening issues across logging media and errors

Exit nonzero when the HTTP server dies first

Add owner ban commands, dependabot PR cap, seed docs

Quarantine models after 3 bad responses until quota day ends

Add owner-only /selftest route-head probe with quota accounting

Fix review findings on limits and stats

Fix stats activity clock and expand diagnostics

Add per-user daily limits with Gemini fallback

setWebhook retries with backoff; fail fast with exit(1) on dead setup (D4: A1-7)

TTS total deadline with shortened notice (D2: A6-4)

Size caps: 20MB Telegram downloads, album semaphore and total, slideshow post budget, honest too-big reply (D1: A6-1, A4-04, A7-1, A6-2, A6-3)

Docs and comments match code: env defaults, routing, rich path, stale notes (A12, A2-5..A2-9, A2-14, A3-5, A6-8, A7-10, A8-07, A5-11)

Single-source limit scalars; generation-checked chat lock acquisition (A5-8, A5-10)

Media lows: safe mime suffix, exact game forms, ext map, WAV check, album skip log, mg cap (A6-5..A6-11, A1-5)

Harden startup/shutdown: token exit, tick survival, off-loop IO, honest command log (A1-1..A1-4, A1-6, A8-06)

Drop unused re-export of media bucket cap (pyflakes)

Validate restored chat state; cap media buckets per chat; stop pruning from dirtying every chat (A5-4, A5-5, A5-6, A5-7, A5-9)

Comment cleanup: short Russian why-comments, no AI boilerplate

Subagent re-review: fix confirmed misses, harden thin tests

Independent re-review: fix missed regressions and close gaps

Fix regressions found by adversarial review: routing, rendering, rate limits

Docs and tests wave 7: routing description matches reality, stronger assertions instead of or-chains, document the deliberate drop_pending_updates tradeoff

Cleanup wave 6: one shared SSE parser and one shared fallback chain, drop dead code and a silent list alias, cover the previously untested failure branches

Medium wave 4: validated env numbers, Upstash backend and shutdown flush tests, deno check/lint in CI, reproducible pip step, docs for missing variables

Security wave 2: gate deploy on push+own repo, 401 and denial logging on admin endpoints, FastAPI schema off, proxy body cap, ownerless pick confined to chat, loud warning about secrets derived from BOT_TOKEN

Fix chat-lock hangs: TTS async with timeout, stream handshake capped, route attempts bounded by remaining budget, one shared chat-lock limit

Claude-parity fixes: callback filters for pick buttons, inline numbered-list splitter in both render paths, prompt ban on glued lists

Luna audit fixes: album lock, multilingual routing, streaming budget, quota and persistence hardening

Media drops are honest: fetchable attachment failure replies error, entry log for media

Ideal routing: voice transcribed into scenario routing, OR qwen newcomer in light chain

Qwen audit fixes: aio Gemini client, snapshot-before-thread state flush, drop dead Fish check, utcnow fix

TikTok UX: localized slideshow status (25 langs), audio duration probe, status lives through music fetch

History summarization: compress overflow into summary instead of silent cut

Menu order: actions draw/tts before settings reset/lang

Groq provider: qwen head + gpt-oss reserve on light routes (calibrated 21.09.2026), streaming, quota/stats/diag, 8 tests

Comment polish: Russian pro-style comments, keep dated evidence short (+426/-2765). Fix: bot.send_chat_action -> bot.bot.send_chat_action (was AttributeError, typing indicator never showed)

Fix prod outage: single bot module namespace + skip fil locale in setMyCommands

Revert guest_message removal (valid Bot API Update field, guest mode) + dependabot grouping

Docs: drop dead dockerignore entry, fix stale test/proxy paths in comments

P2: extract lumen_message_core.py (message handling core) with bot.X re-exports (AUD-A-001)

P2: extract lumen_transport_calls.py (sessions, proxy rotation, tg calls) with bot.X re-exports (AUD-A-001)

P2: extract lumen_rich.py and lumen_media_flow.py with bot.X re-exports (AUD-A-001)

P2: extract lumen_errors.py (model error classification/texts, system prompt) with bot.X re-exports (AUD-A-001)

P2: extract lumen_commands.py (commands, TTS/draw pipelines, pick buttons) with bot.X re-exports (AUD-A-001)

P2: extract lumen_chat_state.py (chat state, quotas, locks, owner utils) with bot.X re-exports (AUD-A-001)

P2: extract lumen_routes.py (OpenRouter/Gemini calls, builders, _run_route) with bot.X re-exports (AUD-A-001)