PromptShip
Migration guide

Migrate from Railway to PromptShip

Railway gets a lot right, persistent containers, real backends, and strong developer ergonomics. Its MCP server and agent surfaces deploy as your Railway account. PromptShip gives each agent its own environment-scoped identity, holds agent-initiated production writes for human approval, records one action ledger, and offers fixed resource tiers.

What changes

  • Predictable tier pricing. Pick a tier, see the price. No per-second usage math, no end-of-month surprise.
  • Governed agent operations. Railway's MCP server acts as your account. PromptShip uses per-agent, environment-scoped tokens, a human approval hold for agent-initiated production writes, and one action ledger.
  • Postgres, Redis, ClickHouse attached natively, not as separate plugins. ClickHouse in particular is rare on competitor platforms.

PromptShip vs Railway

PromptShipRailway
Persistent containers, no cold starts✓✓
Predictable tier pricing (no usage meters)✓✕
Agent identity and environment scopePer-agent, environment-scoped tokenYour Railway account credentials
Agent production writes and attributionHuman approval + one action ledgerNo approval hold or unified agent attribution
Managed Postgres✓✓
Managed Redis-compatible cache✓✓
Managed ClickHouse analytics✓✕
Standard Docker, no lock-in✓✓

Migration steps

  1. 1. Add the PromptShip MCP to your AI IDE.
  2. 2. Connect your GitHub repo, same repo you deployed on Railway.
  3. 3. Attach the databases you need via MCP, Postgres, Redis, ClickHouse.
  4. 4. Point your domain at the new app, verify, decommission Railway.
Get started

Move from Railway to PromptShip

Per-agent controls, production write approvals, one action ledger, and fixed pricing. Free during early access.

[ Join the waitlist ]

Already have access? Install the MCP

$ claude mcp add --transport http promptship https://mcp.promptship.dev/mcp