[3/7] Creating sandbox
  ──────────────────────────────────────────────────
  Sandbox name (lowercase, numbers, hyphens) [my-assistant]: nemoclaw
  Sandbox 'nemoclaw' already exists. Recreate? [y/N]: N
  Keeping existing sandbox.

  [4/7] Configuring inference (NIM)
  ──────────────────────────────────────────────────

  Cloud models:
    1) Nemotron 3 Super 120B (nvidia/nemotron-3-super-120b-a12b)
    2) Kimi K2.5 (moonshotai/kimi-k2.5)
    3) GLM-5 (z-ai/glm5)
    4) MiniMax M2.5 (minimaxai/minimax-m2.5)
    5) Qwen3.5 397B A17B (qwen/qwen3.5-397b-a17b)
    6) GPT-OSS 120B (openai/gpt-oss-120b)

  Choose model [1]: 3
  Using NVIDIA Endpoint API with model: z-ai/glm5

  [5/7] Setting up inference provider
  ──────────────────────────────────────────────────
✓ Created provider nvidia-nim
Gateway inference configured:

  Route: inference.local
  Provider: nvidia-nim
  Model: z-ai/glm5
  Version: 1
  ✓ Inference route set: nvidia-nim / z-ai/glm5

  [6/7] Setting up OpenClaw inside sandbox
  ──────────────────────────────────────────────────
Error:   × status: NotFound, message: "sandbox not found", details: [], metadata: MetadataMap { headers: {"content-type": "application/grpc", "date": "Sat, 21 Mar 2026 10:04:33 GMT"} }

  Command failed (exit 1): cat <<'EOF_NEMOCLAW_SYNC' | openshell sandbox connect "nemoclaw"
set -euo pipefa