Repository Issue Activity (beta)

#85 Add rate limit to outbound calls to external AI

#83 Dependency Issue

#71 --rerank: VLM pass over top-N candidates before trimming

#65 Document the DLQ and `--retry-failed` in the README

#64 Document the `~/.sentrysearch/` cache layout in the README

#78 Add retrieval-accuracy eval (cached y/n judgments → precision@k / MRR across models)

#66 Intel-Mac-specific warning when local backend falls back to CPU

#70 Broader application

#72 Add `--dedupe` to `search` (port from `highlights`)

#63 Add `index --dry-run` to preview chunks and estimated cost

#61 Add `--version` flag to the CLI

#68 Add `sentrysearch benchmark` to collect local-backend hardware reports

#67 `sentrysearch stats` should show all populated backends, not just the auto-detected one

#62 Add `stats --json` for scripting

#53 uv tool install ".[local]" yields "No Solution found when resolving dependencies

#47 Describing and filtering video

#37 Bug: uv tool install on Windows defaults to CPU-only Torch (No GPU detection for local backend)

#38 3rd party API support

#6 [Enhancement] Redesign video overlay to match native Tesla Dashcam UI

#3 Migrate the project from pip to UV