Skip to content

docs: use mounts for examples #91

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
feloy merged 1 commit into from
Jun 17, 2026
+58 −37
Merged

docs: use mounts for examples #91

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
95 changes: 58 additions & 37 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -506,7 +506,7 @@ openshell-image-builder [OPTIONS] <TAG>

### Claude Code agent + Anthropic models provider

```
```sh
$ openshell-image-builder \
--agent claude \
--inference anthropic \
Expand All @@ -525,11 +525,21 @@ $ openshell sandbox create \
--name claude_anthropic_sandbox \
--no-auto-providers \
-- claude

# Or, with podman driver, you can mount the files
# (https://docs.nvidia.com/openshell/reference/sandbox-compute-drivers#podman-driver-config-mounts)
Comment thread
coderabbitai[bot] marked this conversation as resolved.
$ openshell sandbox create \
--from sandbox_image:claude_anthropic \
--provider claude_anthropic_provider \
--driver-config-json '{"podman":{"mounts":[{"type":"bind","source":"/path/to/your/sources","target":"/sandbox/work","read_only":false}]}}' \
--name claude_anthropic_sandbox \
--no-auto-providers \
-- claude
```

### OpenCode agent + Anthropic models provider

```
```sh
$ openshell-image-builder \
--agent opencode \
--inference anthropic \
Expand All @@ -548,45 +558,46 @@ $ openshell sandbox create \
--name opencode_anthropic_sandbox \
--no-auto-providers \
-- opencode

# Or, with podman driver, you can mount the files
# (https://docs.nvidia.com/openshell/reference/sandbox-compute-drivers#podman-driver-config-mounts)
$ openshell sandbox create \
--from sandbox_image:opencode_anthropic \
--provider opencode_anthropic_provider \
--driver-config-json '{"podman":{"mounts":[{"type":"bind","source":"/path/to/your/sources","target":"/sandbox/work","read_only":false}]}}' \
--name opencode_anthropic_sandbox \
--no-auto-providers \
-- opencode
```

### Claude Code agent + Vertex AI models provider

```
```sh
$ openshell-image-builder \
--agent claude \
--inference vertexai \
--model claude-sonnet-4-6 \
sandbox_image:claude_vertexai

# change value of ANTHROPIC_VERTEX_PROJECT_ID and CLOUD_ML_REGION
# also change source paths for mounts
# NOTE: gcloud credentials are not protected with this method
$ openshell sandbox create \
--from sandbox_image:claude_vertexai \
--upload . \
--env CLAUDE_CODE_USE_VERTEX=1 \

@benoitf benoitf Jun 16, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

shouldn't it create first a google-vertex-ai type provider ?
https://docs.nvidia.com/openshell/providers/google-vertex-ai

and then use --provider to give this provider

@feloy feloy Jun 16, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I would prefer to wait for this to stabilize on openshell (NVIDIA/OpenShell#1752 and NVIDIA/OpenShell#1763 are still open) before to document it

--env ANTHROPIC_VERTEX_PROJECT_ID=my-gcp-project \
--env CLOUD_ML_REGION=global \
--driver-config-json '{"podman":{"mounts":[{"type":"bind","source":"/path/to/your/sources","target":"/sandbox/work","read_only":false},{"type":"bind","source":"/path/to/.config/gcloud/application_default_credentials.json","target":"/sandbox/.config/gcloud/application_default_credentials.json","read_only":true}]}}' \
--name claude_vertexai_sandbox \
--no-auto-providers \
--no-tty \
-- bash -c '(\
echo export CLAUDE_CODE_USE_VERTEX=1; \
echo export ANTHROPIC_VERTEX_PROJECT_ID=my-gcp-project; \
echo export CLOUD_ML_REGION=global \
) >> /sandbox/.bashrc'

$ openshell sandbox upload \
claude_vertexai_sandbox \
$HOME/.config/gcloud/application_default_credentials.json \
/sandbox/.config/gcloud/application_default_credentials.json

$ openshell sandbox connect claude_vertexai_sandbox

sandbox:~$ claude
-- claude
```

### OpenCode agent + Ollama (local models)

Ollama must be running on the host before starting the sandbox.

```
```sh
$ openshell-image-builder \
--agent opencode \
--inference ollama \
Expand All @@ -599,6 +610,15 @@ $ openshell sandbox create \
--name opencode_ollama_sandbox \
--no-auto-providers \
-- opencode

# Or, with podman driver, you can mount the files
# (https://docs.nvidia.com/openshell/reference/sandbox-compute-drivers#podman-driver-config-mounts)
$ openshell sandbox create \
--from sandbox_image:opencode_ollama \
--driver-config-json '{"podman":{"mounts":[{"type":"bind","source":"/path/to/your/sources","target":"/sandbox/work","read_only":false}]}}' \
--name opencode_ollama_sandbox \
--no-auto-providers \
-- opencode
```

### OpenCode agent + OpenAI models provider
Expand All @@ -622,6 +642,16 @@ $ openshell sandbox create \
--name opencode_openai_sandbox \
--no-auto-providers \
-- opencode

# Or, with podman driver, you can mount the files
# (https://docs.nvidia.com/openshell/reference/sandbox-compute-drivers#podman-driver-config-mounts)
$ openshell sandbox create \
--from sandbox_image:opencode_openai \
--provider opencode_openai_provider \
--driver-config-json '{"podman":{"mounts":[{"type":"bind","source":"/path/to/your/sources","target":"/sandbox/work","read_only":false}]}}' \
--name opencode_openai_sandbox \
--no-auto-providers \
-- opencode
```

To use an OpenAI-compatible endpoint (e.g. Azure OpenAI, a local proxy, or another provider's API):
Expand All @@ -637,34 +667,25 @@ $ openshell-image-builder \

### OpenCode agent + Vertex AI models provider

```
```sh
$ openshell-image-builder \
--agent opencode \
--inference vertexai \
--model claude-sonnet-4-6 \
sandbox_image:opencode_vertexai

# change value of GOOGLE_CLOUD_PROJECT and VERTEX_LOCATION
# also change source paths for mounts
# NOTE: gcloud credentials are not protected with this method
$ openshell sandbox create \
--from sandbox_image:opencode_vertexai \
--upload . \
--env GOOGLE_APPLICATION_CREDENTIALS=/sandbox/.config/gcloud/application_default_credentials.json \
--env GOOGLE_CLOUD_PROJECT=my-gcp-project \
--env VERTEX_LOCATION=global \
--driver-config-json '{"podman":{"mounts":[{"type":"bind","source":"/path/to/your/sources","target":"/sandbox/work","read_only":false},{"type":"bind","source":"/path/to/.config/gcloud/application_default_credentials.json","target":"/sandbox/.config/gcloud/application_default_credentials.json","read_only":true}]}}' \
--name opencode_vertexai_sandbox \
--no-auto-providers \
--no-tty \
-- bash -c '(\
echo export GOOGLE_CLOUD_PROJECT=my-gcp-project; \
echo export VERTEX_LOCATION=global; \
echo export GOOGLE_APPLICATION_CREDENTIALS=/sandbox/.config/gcloud/application_default_credentials.json \
) >> /sandbox/.bashrc'

$ openshell sandbox upload \
opencode_vertexai_sandbox \
$HOME/.config/gcloud/application_default_credentials.json \
/sandbox/.config/gcloud/application_default_credentials.json

$ openshell sandbox connect opencode_vertexai_sandbox

sandbox:~$ opencode
-- opencode

# select a model with /models
```