docs(skills): make copy of docs-as-skills for Claude

.agents/skills/nemoclaw-configure-inference/SKILL.md

@@ -0,0 +1,85 @@
+---
+name: nemoclaw-configure-inference
+description: Changes the active inference model without restarting the sandbox. Use when change inference runtime, inference routing, openclaw, openshell, switch nemoclaw inference model, switch nemoclaw inference models.
+---
+
+# Nemoclaw Configure Inference
+
+Change the active inference model without restarting the sandbox.
+
+## Prerequisites
+
+- A running NemoClaw sandbox.
+- The OpenShell CLI on your `PATH`.
+
+Change the active inference model while the sandbox is running.
+No restart is required.
+
+## Step 1: Switch to a Different Model
+
+Switching happens through the OpenShell inference route.
+Use the provider and model that match the upstream you want to use.
+
+### NVIDIA Endpoints
+
+```console
+$ openshell inference set --provider nvidia-prod --model nvidia/nemotron-3-super-120b-a12b
+```
+
+### OpenAI
+
+```console
+$ openshell inference set --provider openai-api --model gpt-5.4
+```
+
+### Anthropic
+
+```console
+$ openshell inference set --provider anthropic-prod --model claude-sonnet-4-6
+```
+
+### Google Gemini
+
+```console
+$ openshell inference set --provider gemini-api --model gemini-2.5-flash
+```
+
+### Compatible Endpoints
+
+If you onboarded a custom compatible endpoint, switch models with the provider created for that endpoint:
+
+```console
+$ openshell inference set --provider compatible-endpoint --model <model-name>
+```
+
+```console
+$ openshell inference set --provider compatible-anthropic-endpoint --model <model-name>
+```
+
+If the provider itself needs to change, rerun `nemoclaw onboard`.
+
+## Step 2: Verify the Active Model
+
+Run the status command to confirm the change:
+
+```console
+$ nemoclaw <name> status
+```
+
+Add the `--json` flag for machine-readable output:
+
+```console
+$ nemoclaw <name> status --json
+```
+
+The output includes the active provider, model, and endpoint.
+
+## Step 3: Notes
+
+- The host keeps provider credentials.
+- The sandbox continues to use `inference.local`.
+- Runtime switching changes the OpenShell route. It does not rewrite your stored credentials.
+
+## Related Skills
+
+- `nemoclaw-reference` — Inference Profiles for full profile configuration details

.agents/skills/docs/nemoclaw-get-started/SKILL.md → .agents/skills/nemoclaw-get-started/SKILL.md

@@ -70,7 +70,7 @@ When the install completes, a summary confirms the running environment:
 ```text
 ──────────────────────────────────────────────────
 Sandbox      my-assistant (Landlock + seccomp + netns)
-Model        nvidia/nemotron-3-super-120b-a12b (NVIDIA Endpoint API)
+Model        nvidia/nemotron-3-super-120b-a12b (NVIDIA Endpoints)
 ──────────────────────────────────────────────────
 Run:         nemoclaw my-assistant connect
 Status:      nemoclaw my-assistant status

.agents/skills/docs/nemoclaw-overview/references/how-it-works.md → .agents/skills/nemoclaw-overview/references/how-it-works.md

@@ -93,7 +93,8 @@ After the sandbox starts, the agent runs inside it with all network, filesystem,
 
 Inference requests from the agent never leave the sandbox directly.
 OpenShell intercepts every inference call and routes it to the configured provider.
-NemoClaw routes inference to NVIDIA Endpoints, specifically Nemotron 3 Super 120B through [build.nvidia.com](https://build.nvidia.com). You can switch models at runtime without restarting the sandbox.
+During onboarding, NemoClaw validates the selected provider and model, configures the OpenShell route, and bakes the matching model reference into the sandbox image.
+The sandbox then talks to `inference.local`, while the host owns the actual provider credential and upstream endpoint.
 
 ## Network and Filesystem Policy
 

.agents/skills/docs/nemoclaw-reference/SKILL.md → .agents/skills/nemoclaw-reference/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: nemoclaw-reference
-description: Learns how NemoClaw combines a lightweight CLI plugin with a versioned blueprint to move OpenClaw into a controlled sandbox. Also covers fulls CLI reference for plugin and standalone NemoClaw commands; configurations reference for NVIDIA Endpoint inference profiles. Use when blueprints, cli, inference routing, llms, nemoclaw, nemoclaw architecture, nemoclaw architecture plugin blueprint, nemoclaw cli commands.
+description: Learns how NemoClaw combines a lightweight CLI plugin with a versioned blueprint to move OpenClaw into a controlled sandbox. Also covers fulls CLI reference for plugin and standalone NemoClaw commands; configurations reference for NemoClaw routed inference providers. Use when blueprints, cli, inference routing, llms, nemoclaw, nemoclaw architecture, nemoclaw architecture plugin blueprint, nemoclaw cli commands.
 ---
 
 # Nemoclaw Reference
@@ -11,6 +11,6 @@ Learn how NemoClaw combines a lightweight CLI plugin with a versioned blueprint
 
 - [NemoClaw Architecture — Plugin, Blueprint, and Sandbox Structure](references/architecture.md)
 - [NemoClaw CLI Commands Reference](references/commands.md)
-- [NemoClaw Inference Profiles — NVIDIA Endpoint](references/inference-profiles.md)
+- [NemoClaw Inference Profiles](references/inference-profiles.md)
 - [NemoClaw Network Policies — Baseline Rules and Operator Approval](references/network-policies.md)
 - [NemoClaw Troubleshooting Guide](references/troubleshooting.md)

.agents/skills/docs/nemoclaw-reference/references/commands.md → .agents/skills/nemoclaw-reference/references/commands.md

@@ -24,7 +24,9 @@ Use this command for new installs and for recreating a sandbox after changes to
 $ nemoclaw onboard
 ```
 
-The first run prompts for your NVIDIA API key and saves it to `~/.nemoclaw/credentials.json`.
+The wizard prompts for a provider first, then collects the provider credential if needed.
+Supported non-experimental choices include NVIDIA Endpoints, OpenAI, Anthropic, Google Gemini, and compatible OpenAI or Anthropic endpoints.
+Credentials are stored in `~/.nemoclaw/credentials.json`.
 
 The wizard prompts for a sandbox name.
 Names must follow RFC 1123 subdomain rules: lowercase alphanumeric characters and hyphens only, and must start and end with an alphanumeric character.

.agents/skills/nemoclaw-reference/references/inference-profiles.md

@@ -0,0 +1,64 @@
+# Inference Profiles
+
+NemoClaw configures inference through the OpenShell gateway.
+The agent inside the sandbox talks to `inference.local`, and OpenShell routes that traffic to the provider you selected during onboarding.
+
+## Routed Provider Model
+
+NemoClaw keeps provider credentials on the host.
+The sandbox does not receive your raw OpenAI, Anthropic, Gemini, or NVIDIA API key.
+
+At onboard time, NemoClaw configures:
+
+- an OpenShell provider
+- an OpenShell inference route
+- the baked OpenClaw model reference inside the sandbox
+
+That means the sandbox knows which model family to use, while OpenShell owns the actual provider credential and upstream endpoint.
+
+## Supported Providers
+
+The following non-experimental provider paths are available through `nemoclaw onboard`.
+
+| Provider | Endpoint Type | Notes |
+|---|---|---|
+| NVIDIA Endpoints | OpenAI-compatible | Hosted models on `integrate.api.nvidia.com` |
+| OpenAI | Native OpenAI-compatible | Uses OpenAI model IDs |
+| Other OpenAI-compatible endpoint | Custom OpenAI-compatible | For compatible proxies and gateways |
+| Anthropic | Native Anthropic | Uses `anthropic-messages` |
+| Other Anthropic-compatible endpoint | Custom Anthropic-compatible | For Claude proxies and compatible gateways |
+| Google Gemini | OpenAI-compatible | Uses Google's OpenAI-compatible endpoint |
+
+## Validation During Onboarding
+
+NemoClaw validates the selected provider and model before it creates the sandbox.
+
+- OpenAI-compatible providers:
+  NemoClaw tries `/responses` first, then `/chat/completions`.
+- Anthropic-compatible providers:
+  NemoClaw tries `/v1/messages`.
+- NVIDIA Endpoints manual model entry:
+  NemoClaw also validates the model name against `https://integrate.api.nvidia.com/v1/models`.
+- Compatible endpoint flows:
+  NemoClaw validates by sending a real inference request, because many proxies do not expose a reliable `/models` endpoint.
+
+If validation fails, the wizard does not continue to sandbox creation.
+
+## Local Providers
+
+Local providers use the same routed `inference.local` pattern, but the upstream runtime runs on the host rather than in the cloud.
+
+- Local Ollama
+- Local NVIDIA NIM
+- Local vLLM
+
+Ollama gets additional onboarding help:
+
+- if no models are installed, NemoClaw offers starter models
+- it pulls the selected model
+- it warms the model
+- it validates the model before continuing
+
+## Runtime Switching
+
+For runtime switching guidance, refer to Switch Inference Models (see the `nemoclaw-configure-inference` skill).

.agents/skills/docs/nemoclaw-reference/references/troubleshooting.md → .agents/skills/nemoclaw-reference/references/troubleshooting.md

@@ -139,7 +139,8 @@ Check the active provider and endpoint:
 $ 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.
+If the endpoint is correct but requests still fail, check for network policy rules that may block the connection.
+Then verify the credential and base URL for the provider you selected during onboarding.
 
 ### Agent cannot reach an external host