Skip to content

Roadmap

OneSearch has reached the first stable Docker shape: a single app container with managed Meilisearch by default, plus the legacy external-Meilisearch compose file for people who still want that setup.

The next work is less about proving the idea and more about making big libraries easier to live with.

Shipped

Recent milestones that are already in the product:

  • unified Docker image
  • managed Meilisearch as the default install
  • legacy external Meilisearch compose support
  • setup wizard and JWT auth
  • source path restrictions and read-only mount guidance
  • scheduled indexing with per-source cron
  • full reindex support in the UI, CLI, and API
  • Office, PDF, Markdown, text, code, and config indexing
  • RTF, EPUB, subtitles, CBZ comics, images, RAW photos, and media metadata
  • metadata-only indexing for unsupported files
  • document previews, image previews, RAW embedded previews, and format-aware detail views
  • theme/accent settings and search display settings
  • standalone CLI package

See the changelog for release-by-release details.

Near-term work

These are the kinds of improvements that fit the current product without changing what OneSearch is.

Search polish

  • date, size, and metadata filters
  • better type filtering in the CLI
  • saved searches or recent searches
  • export results as CSV/JSON from the UI
  • more useful empty states when filters hide everything

Indexing and formats

  • archive contents beyond CBZ comics, such as .zip and .tar.gz
  • email formats like .eml and .mbox
  • better handling for very large spreadsheets
  • clearer failed-file cleanup flows
  • more knobs for rich media extraction defaults

Operations

  • cleaner backup/restore workflow
  • better status history for scheduled runs
  • basic metrics endpoint or Prometheus-friendly output
  • easier diagnostics bundle for bug reports

Larger ideas

These are useful, but they need more design before they should be treated as committed work.

Connectors

Cloud and remote storage would probably go through rclone or a similar layer:

  • Google Drive
  • Dropbox
  • OneDrive
  • S3-compatible storage
  • faceted search
  • semantic search with embeddings
  • natural-language filters
  • document summaries
  • auto-tagging

Access control

Right now, if a user can log in, they can search the indexed content. More granular access would need careful design:

  • per-source permissions
  • read-only users
  • SSO/LDAP/OAuth
  • audit logs

Not planned right now

A few things come up naturally, but they are not immediate priorities:

  • mobile apps
  • desktop Electron app
  • browser extension
  • multi-tenant/team workspace features

They may happen later if there is real demand, but the main product still needs the homelab/self-hosted basics to stay solid first.

Suggesting changes

Open a GitHub Discussion for rough ideas, or a GitHub Issue for a specific bug or feature request.

Good requests explain the use case. “Support email archives because I have 20 years of .mbox exports I need to search” is much easier to act on than “add email”.