intent.md

Intent: recreate laju as volt on remix 3 rc baseline

Author: qisthidev. Status: draft.

Problem

Current stack and conventions are tied to the existing implementation, which makes long-term evolution harder across architecture, onboarding, and consistency. If the framework baseline is not pinned, implementation may default to Remix stable v2 and diverge from the intended direction.

Proposed outcome

Recreate the existing project capabilities as Volt on Remix 3.0.0-rc.1 baseline, preserving critical behavior while improving maintainability and architectural clarity. Release target (must-follow): https://github.com/remix-run/remix/releases/tag/remix%403.0.0-rc.1

Affected users and systems

Framework maintainers, application developers using Volt, CI/CD pipelines, SQLite-backed data layer, session/auth subsystem, migrated route modules.

Constraints

Use Remix 3.0.0-rc.1 explicitly (do not fallback to stable v2). No architecture rule violations: Handler → Service → Repository → DB. No SQL outside repositories. No ad-hoc users-table query in auth middleware when session cache is sufficient. Use existing authentication model and avoid introducing unnecessary sensitive data into session payloads. CSRF validation required for all state-changing routes.

Open questions

Will Volt keep SQLite as default-only for v1, or define an official expansion path early? How much UI parity is required in first release vs phased migration? Do we enforce architecture rules via lint/AST checks from day one? Should real-time features ship in v1 scope or after baseline stabilization?