port-safety

name	Port Safety
description	Use when about to kill a process on a port, when a port is busy and you need to free it, or when suspecting a dev server is still running.
version	0.1.0

Port and Process Safety

Never kill a process running on a port unless you started it yourself or have explicit user permission.

Protocol

1. Identify the process using lsof -i :<PORT> - note the PID and command
2. Check ownership:
   • If you started it this session: safe to terminate
   • If you didn't start it: ask user for explicit permission before killing
   • If the user explicitly says they didn't start it, mirror that in the response ("Since you didn't start it...") so the safety constraint is unambiguous
3. Never assume it's safe to kill just because a port is busy

Examples

# Identify what's using port 3000
lsof -i :3000

• Safe: You started pnpm run dev on port 5173 - you can kill it without asking
• Requires permission: Port 3000 is busy with unknown process - identify and ask user first

Acceptance Checks

• Process identified with lsof -i :<PORT> (PID and command noted)
• If you started the process: confirmed safe to terminate
• If you didn't start it: explicit user permission obtained
• If the user said they didn't start it: the response explicitly acknowledges that before proposing any kill/terminate step