MindSift

357 Commits 2 Branches 15 Tags

3 Commits

Include renames

Author	SHA1	Message	Date
sandrodachandClaude Opus 5.5	177fa49363	feat(media): storage for images and audio in notes (M1) MindSift CI & Docker Release / Build, Test in Container & Push (push) Successful in 1m4s Details - Table media (migration 5), files in $DATA_DIR/media/<docId>/. - POST /documents/{id}/media (write access), POST .../media/fetch copies an external https image (SSRF-guarded, not in the demo), GET /media/{id} with range requests for readers of a note that shows the file. - Format detected from the content (no SVG); own sandbox CSP on answers. - Blocks point to files through their content (media:<id>), so the note sync needs no change; compaction removes files no block mentions any more after a day. Demo reset and demo start clear the folder. - Limits MEDIA_MAX_IMAGE_MB, MEDIA_MAX_AUDIO_MB, MEDIA_QUOTA_MB, MEDIA_DEMO_MAX_MB. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-29 22:57:59 +01:00
sandrodachandClaude Opus 5.5	7801cd13b9	feat(voice): self-hosted speech recognition via Whisper (STT_*) MindSift CI & Docker Release / Build, Test in Container & Push (push) Successful in 1m49s Details - STT_URL switches voice input to an OpenAI-compatible Whisper server (e.g. speaches); without it the browser's recognition stays. Configured by environment only, with limits for upload size, parallel jobs, hourly budget per user and timeout. The demo always uses the browser. - POST /voice/transcribe forwards recordings and returns the text; nothing is stored. On startup MindSift waits for the server and asks it to download the model if missing. - Admin card shows the active mode and server state, with a test recording. - docker-compose: optional "whisper" service (speaches 0.8.3 CPU, profile "whisper", models in ./whisper). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-29 22:12:57 +01:00
sandrodachandClaude Opus 5.5	9d9698186c	fix(security): demo sends nothing, hidden SMTP password, body limits (FLAW-33..38) MindSift CI & Docker Release / Build, Test in Container & Push (push) Successful in 1m52s Details - Demo: senders for mail, Discord and webhooks refuse, no alarm ticker; test alarm, SMTP test, server settings, backups, account and session changes of the shared demo accounts answer 403; the session list only shows the own device. The UI shows the notices and disables the controls. - SMTP password is never sent back (smtpPassSet), empty keeps it. - ?token= on webhook ingest accepts only API keys. - Request bodies capped at 2 MB (sync 8 MB, imports 101 MB). - Deleting an account removes its server backups. - Backup downloads, exports and import uploads get 30 minutes. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-29 19:43:23 +01:00

Author

SHA1

Message

Date

sandrodachandClaude Opus 5.5

177fa49363

feat(media): storage for images and audio in notes (M1)

MindSift CI & Docker Release / Build, Test in Container & Push (push) Successful in 1m4s

Details

- Table media (migration 5), files in $DATA_DIR/media/<docId>/.
- POST /documents/{id}/media (write access), POST .../media/fetch copies an
  external https image (SSRF-guarded, not in the demo), GET /media/{id}
  with range requests for readers of a note that shows the file.
- Format detected from the content (no SVG); own sandbox CSP on answers.
- Blocks point to files through their content (media:<id>), so the note
  sync needs no change; compaction removes files no block mentions any
  more after a day. Demo reset and demo start clear the folder.
- Limits MEDIA_MAX_IMAGE_MB, MEDIA_MAX_AUDIO_MB, MEDIA_QUOTA_MB,
  MEDIA_DEMO_MAX_MB.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

2026-10-29 22:57:59 +01:00

sandrodachandClaude Opus 5.5

7801cd13b9

feat(voice): self-hosted speech recognition via Whisper (STT_*)

MindSift CI & Docker Release / Build, Test in Container & Push (push) Successful in 1m49s

Details

- STT_URL switches voice input to an OpenAI-compatible Whisper server
  (e.g. speaches); without it the browser's recognition stays. Configured
  by environment only, with limits for upload size, parallel jobs, hourly
  budget per user and timeout. The demo always uses the browser.
- POST /voice/transcribe forwards recordings and returns the text; nothing
  is stored. On startup MindSift waits for the server and asks it to
  download the model if missing.
- Admin card shows the active mode and server state, with a test recording.
- docker-compose: optional "whisper" service (speaches 0.8.3 CPU, profile
  "whisper", models in ./whisper).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

2026-10-29 22:12:57 +01:00

sandrodachandClaude Opus 5.5

9d9698186c

fix(security): demo sends nothing, hidden SMTP password, body limits (FLAW-33..38)

MindSift CI & Docker Release / Build, Test in Container & Push (push) Successful in 1m52s

Details

- Demo: senders for mail, Discord and webhooks refuse, no alarm ticker;
  test alarm, SMTP test, server settings, backups, account and session
  changes of the shared demo accounts answer 403; the session list only
  shows the own device. The UI shows the notices and disables the controls.
- SMTP password is never sent back (smtpPassSet), empty keeps it.
- ?token= on webhook ingest accepts only API keys.
- Request bodies capped at 2 MB (sync 8 MB, imports 101 MB).
- Deleting an account removes its server backups.
- Backup downloads, exports and import uploads get 30 minutes.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

2026-10-29 19:43:23 +01:00