Storage, Cleanup, and Runtime Visibility for More Reliable Deployments

This release focused on the less glamorous parts of deployment that users feel immediately when something goes wrong: logs, cleanup, persistent storage, and clear recovery paths.

Project Deletion Cleans Up More Reliably

Deleting a project from VibeNest now sends explicit cleanup instructions to Coolify for application config, volumes, Docker cleanup, and connected networks.

The old path relied on the default delete behavior and had very little visibility. If Coolify returned a 404 for an already-orphaned application, the delete could look like an error even though there was nothing left to remove. VibeNest now treats that case as already cleaned up and logs the cleanup path more clearly.

Runtime Logs for Synth Cabal Integrations

Synth Cabal can now request deployment logs through the API, including build logs and runtime logs. This gives AI workers and external automation a way to understand server errors and container crashes without opening a headless browser.

The integration also gained new webhook events for failed deployments, successful deployments, and runtime errors after deployment. Production deployment URLs are now populated correctly, and estimated deploy time is based on recent history instead of a hardcoded value.

Persistent Storage for SQLite Projects

VibeNest now supports persistent Docker volumes for SQLite-style apps. The Storage tab can attach a named volume, keep the .db file across redeploys, and show a warning if volume setup fails.

SQLite detection is intentionally suggestive, not automatic. If a repository looks like it might use SQLite, VibeNest shows a recommendation and lets the user opt in. This avoids attaching unnecessary host volumes to projects that only use SQLite in tests or development scripts.

Safer Postgres Detach

Postgres lifecycle management is more complete. Owners can detach a managed Postgres database with a typed confirmation and an optional final backup.

The order is designed for recoverability: take a final backup, remove app environment variables, restart the app, then delete the database. That avoids leaving an app pointed at a database that has already been destroyed if a restart fails mid-flow.

Old Synth Cabal Binding Columns Removed

The old /bind?token=... flow is gone, so its database columns were removed too. VibeNest now resolves external deployment ownership inline during deployment instead of carrying migration-era binding state.

This does not change the user-facing flow, but it removes a stale concept from admin views and makes the Synth Cabal integration easier to reason about.

What This Enables

The platform is now better at keeping state aligned across VibeNest, Coolify, and the deployed app. That means fewer orphaned containers, fewer mysterious data losses on redeploy, and better signals when a runtime fails after a successful build.