Description
What happened
After onboarding gave me the error from issue #140 I fixed it by using openshell to recreate sandbox with NVidia inference provider (as per comment). However, upon quering the agent the systems is looking for the Anthropic key.
Error sequence
14:05:51 [diagnostic] lane task error: lane=main durationMs=45 error="Error: No API key found for provider "anthropic". Auth store: /sandbox/.openclaw/agents/main/agent/auth-profiles.json (agentDir: /sandbox/.openclaw/agents/main/agent). Configure auth for this agent (openclaw agents add <id>) or copy auth-profiles.json from the main agentDir."
14:05:51 [diagnostic] lane task error: lane=session:agent:main:main durationMs=56 error="Error: No API key found for provider "anthropic". Auth store: /sandbox/.openclaw/agents/main/agent/auth-profiles.json (agentDir: /sandbox/.openclaw/agents/main/agent). Configure auth for this agent (openclaw agents add <id>) or copy auth-profiles.json from the main agentDir."
14:05:51 [model-fallback/decision] model fallback decision: decision=candidate_failed requested=anthropic/claude-opus-4-6 candidate=anthropic/claude-opus-4-6 reason=auth next=none
FailoverError: No API key found for provider "anthropic". Auth store: /sandbox/.openclaw/agents/main/agent/auth-profiles.json (agentDir: /sandbox/.openclaw/agents/main/agent). Configure auth for this agent (openclaw agents add <id>) or copy auth-profiles.json from the main agentDir.
Expected behavior
Sandbox should be using the inference provider specified by openshell command
openshell provider create --name nvidia-nim --type nvidia --credential NVIDIA_API_KEY=
Reproduction Steps
- Install NemoClaw on Ubuntu
- Run
nemoclaw onboard
- Select NVIDIA Cloud API (option 2) at step 4/7
- Accept suggested policy presets (pypi, npm) at step 7/7
- Repeat the step from the comment
openshell sandbox delete my-assistant 2>/dev/null
openshell gateway destroy --name nemoclaw 2>/dev/null
docker volume rm openshell-cluster-nemoclaw 2>/dev/null
openshell gateway start --name nemoclaw
openshell provider create --name nvidia-nim --type nvidia --credential NVIDIA_API_KEY=
openshell inference set --provider nvidia-nim --model nvidia/nemotron-3-super-120b-a12b
openshell sandbox create --name my-assistant --from openclaw
- query openclaw agent from inside the sandbox
openclaw agent --agent main --local -m "Hello, are you running locally?" --session-id test
Environment
- OS: Ubuntu 24.04
- Node.js: v25.8.2
- GPU: NVIDIA GeForce RTX 3050
- Docker version 29.2.1, build a5c7197
- NemoClaw: N/A
$nemoclaw --version
Unknown command: --version
Registered sandboxes: my-assistant
Try: nemoclaw <sandbox-name> connect
Run 'nemoclaw help' for usage.
Debug Output
Logs
`nemoclaw my-assistant status` returns:
$ nemoclaw my-assistant status
Sandbox: my-assistant
Model: nvidia/nemotron-3-super-120b-a12b
Provider: nvidia-nim
GPU: yes
Policies: none
Sandbox:
Id: 3c173e7a-e45c-4fe9-afa3-ce83a0e1fee0
Name: my-assistant
Namespace: openshell
Phase: Ready
Policy:
version: 1
filesystem_policy:
include_workdir: true
read_only:
- /usr
- /lib
- /proc
- /dev/urandom
- /app
- /etc
- /var/log
read_write:
- /sandbox
- /tmp
- /dev/null
landlock:
compatibility: best_effort
process:
run_as_user: sandbox
run_as_group: sandbox
network_policies:
claude_code:
name: claude_code
endpoints:
- host: api.anthropic.com
port: 443
protocol: rest
tls: terminate
enforcement: enforce
access: full
- host: statsig.anthropic.com
port: 443
- host: sentry.io
port: 443
- host: raw.githubusercontent.com
port: 443
- host: platform.claude.com
port: 443
binaries:
- path: /usr/local/bin/claude
- path: /usr/bin/node
github:
name: github
endpoints:
- host: github.com
port: 443
protocol: rest
tls: terminate
enforcement: enforce
rules:
- allow:
method: GET
path: /**/info/refs*
- allow:
method: POST
path: /**/git-upload-pack
binaries:
- path: /usr/bin/git
github_rest_api:
name: github-rest-api
endpoints:
- host: api.github.com
port: 443
protocol: rest
tls: terminate
enforcement: enforce
rules:
- allow:
method: GET
path: /**
- allow:
method: HEAD
path: /**
- allow:
method: OPTIONS
path: /**
binaries:
- path: /usr/local/bin/claude
- path: /usr/bin/gh
gitlab:
name: gitlab
endpoints:
- host: gitlab.com
port: 443
- host: gitlab.mycorp.com
port: 443
binaries:
- path: /usr/bin/glab
nvidia:
name: nvidia
endpoints:
- host: integrate.api.nvidia.com
port: 443
binaries:
- path: /usr/bin/curl
- path: /bin/bash
- path: /usr/local/bin/opencode
nvidia_web:
name: nvidia_web
endpoints:
- host: nvidia.com
port: 443
- host: www.nvidia.com
port: 443
binaries:
- path: /usr/bin/curl
NIM: not runningChecklist
Description
What happened
After onboarding gave me the error from issue #140 I fixed it by using
openshellto recreate sandbox with NVidia inference provider (as per comment). However, upon quering the agent the systems is looking for the Anthropic key.Error sequence
Expected behavior
Sandbox should be using the inference provider specified by openshell command
Reproduction Steps
nemoclaw onboardEnvironment
Debug Output
Logs
Checklist