NVIDIA · ericksoa · Mar 21, 2026 · Mar 20, 2026 · Mar 20, 2026 · Mar 20, 2026
diff --git a/README.md b/README.md
@@ -219,21 +219,7 @@ Run these on the host to set up, connect to, and manage sandboxes.
 | `openshell term`                     | Launch the OpenShell TUI for monitoring and approvals. |
 | `nemoclaw start` / `stop` / `status` | Manage auxiliary services (Telegram bridge, tunnel).   |
 
-### Plugin commands (`openclaw nemoclaw`)
-
-Run these inside the OpenClaw CLI. These commands are under active development and may not all be functional yet.
-
-| Command                                    | Description                                              |
-|--------------------------------------------|----------------------------------------------------------|
-| `openclaw nemoclaw launch [--profile ...]` | Bootstrap OpenClaw inside an OpenShell sandbox.          |
-| `openclaw nemoclaw status`                 | Show sandbox health, blueprint state, and inference.     |
-| `openclaw nemoclaw logs [-f]`              | Stream blueprint execution and sandbox logs.             |
-
-See the full [CLI reference](https://docs.nvidia.com/nemoclaw/latest/reference/commands.html) for all commands, flags, and options.
-
-> **Known limitations:**
-> - The `openclaw nemoclaw` plugin commands are under active development. Use the `nemoclaw` host CLI as the primary interface.
-> - Setup may require manual workarounds on some platforms. File an issue if you encounter blockers.
+See the full [CLI reference](https://docs.nvidia.com/nemoclaw/latest/reference/commands.md) for all commands, flags, and options.
 
 ---
 

diff --git a/docs/about/how-it-works.md b/docs/about/how-it-works.md
@@ -75,7 +75,7 @@ Thin plugin, versioned blueprint
 : The plugin stays small and stable. Orchestration logic lives in the blueprint and evolves on its own release cadence.
 
 Respect CLI boundaries
-: The `nemoclaw` CLI is the primary interface. Plugin commands are available under `openclaw nemoclaw` but do not override built-in OpenClaw commands.
+: The `nemoclaw` CLI is the primary interface for sandbox management.
 
 Supply chain safety
 : Blueprint artifacts are immutable, versioned, and digest-verified before execution.
@@ -91,7 +91,7 @@ Reproducible setup
 
 NemoClaw is split into two parts:
 
-- The *plugin* is a TypeScript package that powers the `nemoclaw` CLI and also registers commands under `openclaw nemoclaw`.
+- The *plugin* is a TypeScript package that registers an inference provider and the `/nemoclaw` slash command inside the sandbox.
   It handles user interaction and delegates orchestration work to the blueprint.
 - The *blueprint* is a versioned Python artifact that contains all the logic for creating sandboxes, applying policies, and configuring inference.
   The plugin resolves, verifies, and executes the blueprint as a subprocess.

diff --git a/docs/inference/switch-inference-providers.md b/docs/inference/switch-inference-providers.md
@@ -44,13 +44,13 @@ The `nemoclaw onboard` command stores this key in `~/.nemoclaw/credentials.json`
 Run the status command to confirm the change:
 
 ```console
-$ openclaw nemoclaw status
+$ nemoclaw <name> status
 ```
 
 Add the `--json` flag for machine-readable output:
 
 ```console
-$ openclaw nemoclaw status --json
+$ nemoclaw <name> status --json
 ```
 
 The output includes the active provider, model, and endpoint.

diff --git a/docs/monitoring/monitor-sandbox-activity.md b/docs/monitoring/monitor-sandbox-activity.md
@@ -32,13 +32,7 @@ Use the NemoClaw status, logs, and TUI tools together to inspect sandbox health,
 Run the status command to view the sandbox state, blueprint run information, and active inference configuration:
 
 ```console
-$ openclaw nemoclaw status
-```
-
-For machine-readable output, add the `--json` flag:
-
-```console
-$ openclaw nemoclaw status --json
+$ nemoclaw <name> status
 ```
 
 Key fields in the output include the following:
@@ -47,33 +41,22 @@ Key fields in the output include the following:
 - Blueprint run ID, which is the identifier for the most recent blueprint execution.
 - Inference provider, which shows the active provider, model, and endpoint.
 
-If you run `openclaw nemoclaw status` from inside the sandbox, the command detects the sandbox context and reports it. Host-level sandbox and inference details are not available from within the sandbox. Run `openshell sandbox list` on the host to check the underlying sandbox state.
+Run `nemoclaw <name> status` on the host to check sandbox state. Use `openshell sandbox list` for the underlying sandbox details.
 
 ## View Blueprint and Sandbox Logs
 
 Stream the most recent log output from the blueprint runner and sandbox:
 
 ```console
-$ openclaw nemoclaw logs
+$ nemoclaw <name> logs
 ```
 
 To follow the log output in real time:
 
 ```console
-$ openclaw nemoclaw logs -f
+$ nemoclaw <name> logs -f
 ```
 
-To display a specific number of log lines:
-
-```console
-$ openclaw nemoclaw logs -n 100
-```
-
-To view logs for a specific blueprint run instead of the most recent one:
-
-```console
-$ openclaw nemoclaw logs --run-id <id>
-```
 
 ## Monitor Network Activity in the TUI
 
@@ -104,8 +87,8 @@ $ openclaw agent --agent main --local -m "Test inference" --session-id debug
 
 If the request fails, check the following:
 
-1. Run `openclaw nemoclaw status` to confirm the active provider and endpoint.
-2. Run `openclaw nemoclaw logs -f` to view error messages from the blueprint runner.
+1. Run `nemoclaw <name> status` to confirm the active provider and endpoint.
+2. Run `nemoclaw <name> logs -f` to view error messages from the blueprint runner.
 3. Verify that the inference endpoint is reachable from the host.
 
 ## Related Topics

diff --git a/docs/network-policy/customize-network-policy.md b/docs/network-policy/customize-network-policy.md
@@ -62,7 +62,7 @@ The wizard picks up the modified policy file and applies it to the sandbox.
 Check that the sandbox is running with the updated policy:
 
 ```console
-$ openclaw nemoclaw status
+$ nemoclaw <name> status
 ```
 
 ## Dynamic Changes

diff --git a/docs/reference/architecture.md b/docs/reference/architecture.md
@@ -24,8 +24,8 @@ NemoClaw has two main components: a TypeScript plugin that integrates with the O
 
 ## NemoClaw Plugin
 
-The plugin is a thin TypeScript package that registers commands under `openclaw nemoclaw`.
-It runs in-process with the OpenClaw gateway and handles user-facing CLI interactions.
+The plugin is a thin TypeScript package that registers an inference provider and the `/nemoclaw` slash command.
+It runs in-process with the OpenClaw gateway inside the sandbox.
 
 ```text
 nemoclaw/

diff --git a/docs/reference/commands.md b/docs/reference/commands.md
@@ -20,77 +20,7 @@ status: published
 
 # Commands
 
-NemoClaw provides two command interfaces.
-The plugin commands run under the `openclaw nemoclaw` namespace inside the OpenClaw CLI.
-The standalone `nemoclaw` binary handles host-side setup, deployment, and service management.
-Both interfaces are installed when you run `npm install -g nemoclaw`.
-
-## Plugin Commands
-
-### `openclaw nemoclaw launch`
-
-Bootstrap OpenClaw inside an OpenShell sandbox.
-If NemoClaw detects an existing host installation, `launch` stops unless you pass `--force`.
-
-```console
-$ openclaw nemoclaw launch [--force] [--profile <profile>]
-```
-
-`--force`
-: Skip the ergonomics warning and force plugin-driven bootstrap. Without this flag,
-  NemoClaw recommends using `openshell sandbox create` directly for new installs.
-
-`--profile <profile>`
-: Blueprint profile to use. Default: `default`.
-
-### `nemoclaw <name> connect`
-
-Open an interactive shell inside the OpenClaw sandbox.
-Use this after launch to connect and chat with the agent through the TUI or CLI.
-
-```console
-$ nemoclaw my-assistant connect
-```
-
-If the TUI view is not a good fit for very long responses, use the CLI form instead:
-
-```console
-$ openclaw agent --agent main --local -m "<prompt>" --session-id <id>
-```
-
-This is the recommended workaround when you need the full response printed directly in the terminal.
-
-### `openclaw nemoclaw status`
-
-Display sandbox health, blueprint run state, and inference configuration.
-
-```console
-$ openclaw nemoclaw status [--json]
-```
-
-`--json`
-: Output as JSON for programmatic consumption.
-
-When running inside an active OpenShell sandbox, the status command detects the sandbox context and reports "active (inside sandbox)" instead of false negatives.
-Host-side sandbox state and inference configuration are not inspectable from inside the sandbox.
-Run `openshell sandbox list` on the host to check the underlying sandbox state.
-
-### `openclaw nemoclaw logs`
-
-Stream blueprint execution and sandbox logs.
-
-```console
-$ openclaw nemoclaw logs [-f] [-n <count>] [--run-id <id>]
-```
-
-`-f, --follow`
-: Follow log output, similar to `tail -f`.
-
-`-n, --lines <count>`
-: Number of lines to show. Default: `50`.
-
-`--run-id <id>`
-: Show logs for a specific blueprint run instead of the latest.
+The `nemoclaw` CLI is the primary interface for managing NemoClaw sandboxes. It is installed when you run `npm install -g nemoclaw`.
 
 ### `/nemoclaw` Slash Command
 

diff --git a/docs/reference/troubleshooting.md b/docs/reference/troubleshooting.md
@@ -151,7 +151,7 @@ Run `nemoclaw onboard` to recreate the sandbox from the same blueprint and polic
 ### Status shows "not running" inside the sandbox
 
 This is expected behavior.
-When running `openclaw nemoclaw status` inside an active sandbox, host-side sandbox state and inference configuration are not inspectable.
+When checking status inside an active sandbox, host-side sandbox state and inference configuration are not inspectable.
 The status command detects the sandbox context and reports "active (inside sandbox)" instead.
 
 Run `openshell sandbox list` on the host to check the underlying sandbox state.
@@ -162,7 +162,7 @@ Verify that the inference provider endpoint is reachable from the host.
 Check the active provider and endpoint:
 
 ```console
-$ openclaw nemoclaw status
+$ nemoclaw <name> status
 ```
 
 If the endpoint is correct but requests still fail, check for network policy rules that may block the connection, and verify that your NVIDIA API key is valid.
@@ -184,8 +184,7 @@ Refer to [Customize the Network Policy](../network-policy/customize-network-poli
 View the error output for the failed blueprint run:
 
 ```console
-$ openclaw nemoclaw logs --run-id <id>
+$ nemoclaw <name> logs
 ```
 
-If the run ID is unknown, omit `--run-id` to view logs from the most recent run.
 Use `--follow` to stream logs in real time while debugging.
diff --git a/nemoclaw-blueprint/policies/openclaw-sandbox.yaml b/nemoclaw-blueprint/policies/openclaw-sandbox.yaml
@@ -10,7 +10,7 @@
 #   strict  — this file. Minimum for onboard + basic agent operation.
 #   relaxed — adds third-party model providers, broader web access.
 #
-# To add endpoints: update this file and re-run `openclaw nemoclaw migrate`
+# To add endpoints: update this file and re-run `nemoclaw onboard`
 # or apply dynamically via `openshell policy set`.
 
 version: 1

diff --git a/nemoclaw/src/blueprint/exec.ts b/nemoclaw/src/blueprint/exec.ts