Add ship-* Claude skills and plan

Introduces a skills/ directory with 8 Claude skills that reimagine ship
as a set of composable, human-driven deployment tools backed by Claude's
reasoning rather than a rigid CLI.

Skills:
- ship-setup: one-time VPS config, saves host to ~/.config/ship/config.json
- ship-status: derives live state from server, no local state file
- ship-env: read/write env vars with merge semantics, never overwrites
- ship-binary: deploy Go binaries with SQLite backup, correct restart behavior
- ship-caddy: manage per-app Caddyfile with validate-before-reload
- ship-service: systemd management and log inspection
- ship-static: rsync static sites with SPA routing support
- ship-deploy: orchestration runbook tying the others together

Also adds SKILLS_PLAN.md documenting the architecture and rationale.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

1 files changed, 133 insertions, 0 deletions

diff --git a/skills/ship-service/SKILL.md b/skills/ship-service/SKILL.md
new file mode 100644
index 0000000..e4d7510
--- /dev/null
+++ b/skills/ship-service/SKILL.md
@@ -0,0 +1,133 @@
+---
+name: ship-service
+description: Manage systemd services for deployed apps. Start, stop, restart, view status and logs. Use when you need to control a running service or diagnose problems.
+argument-hint: "<app-name> [start|stop|restart|status|logs] [host-nickname]"
+---
+
+# ship-service
+
+Manage systemd services for apps deployed on a ship VPS.
+
+## Read Config
+
+```bash
+python3 -c "
+import json, os
+cfg = json.load(open(os.path.expanduser('~/.config/ship/config.json')))
+nick = '<nickname-or-default>'
+h = cfg['hosts'].get(nick, cfg['hosts'][cfg['default']])
+print(h['host'])
+"
+```
+
+## Usage Patterns
+
+### Status
+
+```bash
+ssh <host> "sudo systemctl status <app-name> --no-pager"
+```
+
+Shows whether the service is running, its PID, memory usage, and recent log lines.
+
+### Restart
+
+```bash
+ssh <host> "sudo systemctl restart <app-name>"
+```
+
+Use after changing env vars or replacing the binary.
+
+### Stop
+
+```bash
+ssh <host> "sudo systemctl stop <app-name>"
+```
+
+### Start
+
+```bash
+ssh <host> "sudo systemctl start <app-name>"
+```
+
+### View logs
+
+Recent logs (last 50 lines):
+```bash
+ssh <host> "sudo journalctl -u <app-name> -n 50 --no-pager"
+```
+
+Follow live logs:
+```bash
+ssh <host> "sudo journalctl -u <app-name> -f"
+```
+
+Logs since last boot:
+```bash
+ssh <host> "sudo journalctl -u <app-name> -b --no-pager"
+```
+
+### Enable (start on boot)
+
+```bash
+ssh <host> "sudo systemctl enable <app-name>"
+```
+
+### Disable (don't start on boot)
+
+```bash
+ssh <host> "sudo systemctl disable <app-name>"
+```
+
+### View the systemd unit file
+
+```bash
+ssh <host> "sudo cat /etc/systemd/system/<app-name>.service"
+```
+
+### Remove a service entirely
+
+Only do this if the user explicitly asks to remove/uninstall an app:
+
+```bash
+ssh <host> "sudo systemctl stop <app-name> && sudo systemctl disable <app-name> && sudo rm /etc/systemd/system/<app-name>.service && sudo systemctl daemon-reload"
+```
+
+Confirm with the user before removing. Note that this does not delete the binary,
+data directory, env file, or Caddy config — those are managed separately.
+
+## Diagnosing Problems
+
+If a service is failing, check:
+
+1. Service status for the error:
+```bash
+ssh <host> "sudo systemctl status <app-name> --no-pager"
+```
+
+2. Full logs for context:
+```bash
+ssh <host> "sudo journalctl -u <app-name> -n 100 --no-pager"
+```
+
+3. Whether the binary exists and is executable:
+```bash
+ssh <host> "ls -la /usr/local/bin/<app-name>"
+```
+
+4. Whether the env file exists and looks correct:
+```bash
+ssh <host> "sudo cat /etc/ship/env/<app-name>.env"
+```
+
+5. Whether the port is already in use by something else:
+```bash
+ssh <host> "sudo ss -tlnp | grep <port>"
+```
+
+## Notes
+
+- If the user just says "restart foodtracker" or "check the logs for myapp", infer the action
+- After a restart, give it a moment then check status to confirm it came up
+- If a service repeatedly crashes, look at the logs before suggesting a fix
+- Use default host unless another is specified