Address CodeRabbit feedback:

1. Remove early return when gateway is running — always run
   recoverDashboardChain() so broken forward/CORS links get repaired
   even when the in-sandbox gateway is healthy (#2042, #1178).

2. Forward port param through restartGateway → recoverSandboxProcesses
   so the fallback recovery script uses the chain's port instead of
   hardcoded DASHBOARD_PORT.

The wasRunning field now reflects the actual gateway probe result
regardless of whether recovery was attempted.

Refs: #2390

…ailures

The E2E tests (sandbox-survival-e2e, skip-permissions-e2e) failed with
exit code 124 (timeout hang) because recoverDashboardChain immediately
re-verified after restarting the gateway — the gateway HTTP listener
hadn't bound yet, so verifyDashboardChain reported it unhealthy.

Add sleepSeconds(3) after a successful restartGateway call to give the
gateway time to bind its HTTP port before the chain re-verification.
This matches the original recovery code's behavior (sleepSeconds(3)
between recoverSandboxProcesses and isSandboxGatewayRunning).

Also removes a redundant type annotation in downloadSandboxConfig.

Refs: #2390

…port

Address Aaron's review feedback:

1. Port derivation: use registry.getSandbox(name)?.dashboardPort instead
   of agent?.forwardPort to correctly recover multi-sandbox setups with
   auto-allocated or user-overridden ports (e.g. 18790).

2. Agent gating: only run full dashboard chain recovery (CORS check via
   /sandbox/.openclaw/openclaw.json) for OpenClaw sandboxes. Non-OpenClaw
   agents (Hermes) use different config paths and don't expose the
   OpenClaw control UI — fall back to gateway-only recovery via new
   checkAndRecoverGatewayOnly() helper.

The gateway-only path preserves the original pre-#2398 behavior for
Hermes: check gateway → restart if dead → re-forward → done.

Refs: #2390

…ate (#2701)

After a pod recreate, the sandbox's /tmp is fresh and the NODE_OPTIONS
preload guard chain (safety-net, ciao-network-guard, http-proxy-fix,
nemotron-fix, slack-channel-guard, slack-token-rewriter) and the
proxy-env.sh aggregator are all gone. The recovery path previously
detected this and warned but launched the gateway anyway — causing a
crash-loop from @homebridge/ciao's os.networkInterfaces() call.

Fix:
- Extract guard JS files into scripts/lib/guards/ as standalone files
- Add scripts/lib/emit-guards.sh that installs guards to /tmp with
  correct permissions (root:root 444) and generates proxy-env.sh
- Install both into the Docker image at /usr/local/lib/nemoclaw/
- Add src/lib/guard-recovery.ts with checkGuardsPresent() and
  reEmitGuards() — uses kubectl exec (via docker exec) to run
  emit-guards.sh as root inside the sandbox, bypassing Landlock
- Wire guard re-emission into recoverSandboxProcesses() before
  the gateway launch script runs

The recovery path now: detect missing guards → re-emit via kubectl exec
→ source the freshly-written proxy-env.sh → launch gateway with guards.

Refs: #2701, #2478