Commit graph

3 commits

Author SHA1 Message Date
917954d12d Sign-in lockout with console unlock, and forgotten-password reset
All checks were successful
Verify and publish V2 container / build-and-push (push) Successful in 29s
- Wrong passwords in a row are counted across every device and kept in
  the database. The first 3 are free; from the 3rd the next attempt
  waits 15 s, 30 s, 1, 2, 5, 15 and 30 minutes; the 10th locks sign-in
  until the unlock code printed to the server log (and saved to
  unlock-code.txt) is entered. Signed-in devices keep working. Only one
  password check runs at a time, so parallel guesses cannot skip a wait.
- "Forgot password?" prints a reset code to the log (reset-code.txt),
  valid once for 15 minutes. It sets a new password, signs out every
  device, lifts any lockout and signs in. Asking again while a code is
  valid prints nothing new.
- Codes are stored hashed; wrong codes count toward the per-address
  limit. RESET_ADMIN_PASSWORD also lifts a lockout.
- Sign-in page: a live countdown during a wait, an unlock step and a
  reset step.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 15:46:16 -07:00
f0aabe4d5d First revision: DeepSeek, numeric questions, background assistant, UI fixes
All checks were successful
Verify and publish V2 container / build-and-push (push) Successful in 1m2s
Providers and pricing
- Add a DeepSeek provider: thinking mode with low/high/max effort, reasoning_content
  replayed between tool steps, cache-hit usage, /models discovery and catalog prices.
- Fix "does not support tool calling": a rejected forced tool_choice (DeepSeek thinking
  mode) now asks for the call in the prompt; only a genuine no-tools 400 is "unsupported".
- Seven reasoning-effort levels (Off to Max) mapped per adapter.
- Cached-input price per model, cache-write and DeepSeek off-peak aware cost estimates.
- Built-in bge-small embedding model (transformers.js) as the semantic search default.

Quizzes
- Numeric (type-the-answer) questions: units ignored, optional ± tolerance; editor,
  runner, importer, templates and generation prompts.
- Quiz start screen instead of showing answers; navigator ring no longer clipped and
  key legend removed; clock only for timed exams; 7-point letter grades.

Assistant and generation
- Assistant turns run on the server independently of the browser, with re-attach and
  an explicit stop; drafts tray shows generations in progress.
- Per-chat model switcher, @-tagging of class materials, full-width assistant page and
  a smaller assistant button.
- Editable generation instructions and built-in presets, separate from the chatbot
  import instructions.

Other
- Dialog footer padding, keyboard hints removed, copy buttons work over plain HTTP,
  class and folder shares include documents.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-15 12:48:17 -07:00
eae109c166 Add HTTP routes, app wiring, V1 migration, backups, templates and API tests
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
2026-11-11 08:22:47 -07:00