# Deploy Dev > Start the local development environment (Docker DB + Redis, native API + Vite) Use when this capability is needed. - Skill: `tomevault-io/deploy-dev-2` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/deploy-dev-2` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/deploy-dev-2/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: DevOps & Infra - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/deploy-dev-2 --- # Deploy Dev Environment **Remote-first (default when `RL_PROXMOX_HOST` is reachable):** spin a per-claim allinone env on the rl-infra VM. This gives a prod-like (supervisor + nginx + unix-socket Redis) stack instead of the dev watch-mode loop and offloads CPU/RAM from the laptop: ```bash rl claim --branch $(git branch --show-current) # may enqueue with queue_position=N # If the CLI prints `enqueued queue_position=N`, fleet contention is real — # call `rl_claim_wait` (MCP) until head, OR run locally via `RL_TARGET=local ./scripts/deploy_dev.sh --ci`. rl env spin $(git branch --show-current | tr / -) ``` The output prints the URL (`https://.rl.lan`) and the per-env Postgres container name. `rl env destroy ` tears it down; the gc-sweeper auto-prunes after 24h. See `rl-infra/README.md` and `.claude/skills/_shared/rl-infra-fleet.md`. **Local fallback (airplane mode / `RL_TARGET=local`):** start the local dev environment using the deploy script. Run: `./scripts/deploy_dev.sh` This starts: - Docker containers for PostgreSQL and Redis - NestJS API in watch mode on http://localhost:3000 - Vite dev server with HMR on http://localhost:5173 After startup, display the admin credentials and available URLs from the script output. --- > Source: [sjdodge123/Raid-Ledger](https://github.com/sjdodge123/Raid-Ledger) — distributed by [TomeVault](https://tomevault.io).