| name | railway |
| description | Manage the Railway-hosted proxy server. Use when the user says "railway", "deploy proxy", "proxy logs", "proxy status", "redeploy", or wants to check/manage the Railway deployment. |
Railway Proxy Management
Manage the overhead-tracker-proxy service on Railway (project: resourceful-integrity).
Platform Detection
uname -s
- Darwin → prefix railway/brew commands with
eval "$(/opt/homebrew/bin/brew shellenv)" &&
- MINGW/MSYS/CYGWIN → run commands directly (railway CLI must be in PATH)
Service info
- Project: resourceful-integrity
- Service: overhead-tracker-proxy
- Custom domain: api.overheadtracker.com
- Volume: /data (route cache + reports)
Commands
Parse the user's intent and run the appropriate command(s) below.
Deploy (user says "deploy", "push", "update proxy", "railway up")
- Syntax-check:
node --check server/server.js
- Deploy (MUST run from project root, NOT from server/):
railway up --service overhead-tracker-proxy
- Wait 30 seconds, then verify:
curl -s https://api.overheadtracker.com/status | head -c 300
Status (user says "status", "check proxy", "is proxy up")
curl -s https://api.overheadtracker.com/status
Logs (user says "logs", "proxy logs", "what happened")
railway service logs --service overhead-tracker-proxy
Redeploy (user says "redeploy", "restart proxy")
railway service redeploy --service overhead-tracker-proxy --yes
Then wait 30 seconds and verify with /status.
Variables (user says "env", "variables", "config")
railway variables
Set variable (user says "set VAR=value")
railway variables set KEY=VALUE
Report summary
Tell the user:
- Whether the action succeeded
- The /status response if applicable
- Any errors encountered