NVIDIA · miyoungc · Jul 29, 2026 · Jul 29, 2026
diff --git a/.agents/skills/nemoclaw-contributor-update-docs/SKILL.md b/.agents/skills/nemoclaw-contributor-update-docs/SKILL.md
@@ -117,6 +117,11 @@ For each relevant commit, determine which doc page(s) it affects. Use this mappi
 | `Dockerfile` | `docs/reference/architecture.mdx` |
 | Inference-related changes | Start at `docs/inference/how-inference-routing-works.mdx`, then update the focused provider, model, setup, management, or validation page that owns the behavior. |
 
+For every target page, determine which agent runtimes execute the behavior and which guide variants must publish it.
+Use implementation gates, tests, or accepted product scope as evidence for each inclusion or exclusion.
+Do not infer applicability from the page's current navigation placement.
+Host-side NemoClaw and OpenShell behavior is not OpenClaw-only unless the implementation or accepted scope establishes that restriction.
+
 If a commit does not map to any existing page but introduces a user-visible concept, flag it as needing a new page.
 If a commit already changes files under `docs/`, include those pages in the target page list and run a docs review or edit pass against them using the style guidance in Step 5.
 Do not assume an existing doc change is complete, correctly placed, or style-compliant just because it landed with the source commit.
@@ -201,7 +206,9 @@ When updating an existing page:
 When creating a new page:
 
 - Follow the frontmatter template from existing pages in `docs/`.
-- Add the page to the appropriate navigation entry in `docs/index.yml`.
+- Add the page to every applicable navigation variant in `docs/index.yml`.
+- Use generated navigation targets when the page applies to more than one variant.
+- Declare the exact `agent-variants` subset in frontmatter when the page intentionally applies to fewer than OpenClaw, Hermes, and Deep Agents.
 
 ## Step 6: Present the Results
 

diff --git a/docs/AGENTS.md b/docs/AGENTS.md
@@ -24,6 +24,9 @@ Treat `docs/` as the source of truth for published content and AI-agent Markdown
 - Check `docs/.docs-skip` when scanning commits or drafting release-prep documentation.
 - Read the full target page before editing it.
 - Map code changes to existing pages before proposing a new page.
+- For every target page, determine which agent runtimes execute the documented behavior and which guide variants must publish it.
+- Use source code, tests, or accepted product scope as evidence for each inclusion or exclusion.
+- Do not infer agent applicability from the page's current navigation placement.
 - Update `.agents/skills/nemoclaw-user-guide/SKILL.md` only when AI-agent docs routing guidance changes.
 
 ## Writing Rules
@@ -44,8 +47,10 @@ Treat `docs/` as the source of truth for published content and AI-agent Markdown
 
 ## NemoClaw Doc Patterns
 
-- Use `$$nemoclaw` for host CLI command examples on shared OpenClaw, Hermes, and Deep Agents pages.
-- Use literal command names on pages that have only one agent variant.
+- Use `$$nemoclaw` for host CLI command examples on source pages shared by OpenClaw, Hermes, and Deep Agents guide variants.
+- Use literal command names on source pages published for one guide variant.
+- Publish shared source pages through generated navigation targets in every applicable guide variant.
+- Declare `agent-variants` in frontmatter when a source page intentionally applies to fewer than all three guide variants.
 - Use `<AgentOnly>` blocks only when content differs by behavior, setup flow, state layout, or agent-specific wording.
 - Treat `<AgentOnly>` as a non-nested build-time directive with opening and closing tags at the first column on their own lines; do not import a runtime component for it.
 - Use route-style links without `.mdx` extensions for links between docs pages.

diff --git a/docs/CONTRIBUTING.md b/docs/CONTRIBUTING.md
@@ -171,6 +171,12 @@ The source pages stay in their normal `docs/` locations, and generated pages are
 Navigation in `docs/index.yml` points Fern at generated pages for shared entries so Fern still renders normal fenced code blocks with copy buttons and syntax highlighting.
 OpenClaw-only, Hermes-only, or Deep Agents-only pages stay as source pages in navigation.
 
+Determine page applicability from the implementation, tests, or accepted product scope before adding or moving navigation entries.
+Do not use the current navigation tree as evidence that a page is agent-specific.
+Publish a shared source page through generated navigation targets in every applicable variant.
+When a page intentionally applies to fewer than all three variants, declare the exact subset in frontmatter, for example `agent-variants: ["openclaw", "hermes"]`.
+The sync command fails when a subset declaration is missing or differs from navigation membership.
+
 When shared page content is the same except for the host CLI binary, write one source page and use `$$nemoclaw` as a build-time placeholder.
 Do not duplicate fenced code blocks or inline command examples only to switch among `nemoclaw`,
 `nemohermes`, and `nemo-deepagents`.
@@ -262,6 +268,12 @@ position: 1
 ---
 ```
 
+When the page intentionally applies to fewer than OpenClaw, Hermes, and Deep Agents, add the exact subset to frontmatter:
+
+```yaml
+agent-variants: ["openclaw", "hermes"]
+```
+
 ### Page Structure
 
 1. Start MDX pages with a one- or two-sentence introduction stating what the page covers.

diff --git a/docs/about/ecosystem-deepagents.mdx b/docs/about/ecosystem-deepagents.mdx
@@ -8,6 +8,7 @@ description-agent: "Explains how LangChain Deep Agents Code, OpenShell, and Nemo
 keywords: ["nemoclaw deep agents ecosystem", "dcode openshell sandbox", "nemo-deepagents", "langchain deep agents code"]
 content:
   type: "concept"
+agent-variants: ["deepagents"]
 ---
 NemoClaw provides onboarding, lifecycle management, policy, and inference routing for LangChain Deep Agents Code in OpenShell containers.
 Use the `nemo-deepagents` CLI alias when you work from the Deep Agents guide; it is equivalent to `nemoclaw` with `langchain-deepagents-code` pre-selected.

diff --git a/docs/about/ecosystem-hermes.mdx b/docs/about/ecosystem-hermes.mdx
@@ -8,6 +8,7 @@ description-agent: "Explains how Hermes, OpenShell, and NemoClaw form the ecosys
 keywords: ["nemoclaw ecosystem", "hermes agent", "nemohermes", "nemoclaw vs openshell", "run hermes openshell sandbox"]
 content:
   type: "concept"
+agent-variants: ["hermes"]
 ---
 NemoClaw provides onboarding, lifecycle management, and Hermes operations in OpenShell containers.
 Use the `nemohermes` CLI alias when you work from the Hermes agent guide; it is equivalent to `nemoclaw` with the Hermes agent pre-selected.

diff --git a/docs/about/ecosystem.mdx b/docs/about/ecosystem.mdx
@@ -8,6 +8,7 @@ description-agent: "Explains how OpenClaw, OpenShell, and NemoClaw form the ecos
 keywords: ["nemoclaw ecosystem", "openclaw openshell", "nemoclaw vs openshell", "sandboxed openclaw"]
 content:
   type: "concept"
+agent-variants: ["openclaw"]
 ---
 NemoClaw provides onboarding, lifecycle management, and OpenClaw operations in OpenShell containers.
 

diff --git a/docs/configure-agents/configure-agent-heartbeats.mdx b/docs/configure-agents/configure-agent-heartbeats.mdx
@@ -8,6 +8,7 @@ description-agent: "Configures OpenClaw heartbeat cadence through NemoClaw onboa
 keywords: ["nemoclaw heartbeat", "openclaw heartbeat", "HEARTBEAT.md"]
 content:
   type: "how_to"
+agent-variants: ["openclaw"]
 ---
 
 OpenClaw heartbeats run periodic main-session agent turns.

diff --git a/docs/configure-agents/understand-context-compaction.mdx b/docs/configure-agents/understand-context-compaction.mdx
@@ -8,6 +8,7 @@ description-agent: "Explains OpenClaw context compaction behavior in NemoClaw-ma
 keywords: ["nemoclaw context compaction", "openclaw transcript compaction", "openclaw context management"]
 content:
   type: "concept"
+agent-variants: ["openclaw"]
 ---
 
 NemoClaw configures safeguard compaction for OpenClaw sandboxes that use its managed `inference.local` route, except for Local Ollama routes.

diff --git a/docs/deployment/gateway-lifecycle-authority.mdx b/docs/deployment/gateway-lifecycle-authority.mdx
@@ -9,6 +9,10 @@ keywords: ["nemoclaw gateway lifecycle", "externally supervised gateway", "opens
 content:
   type: "reference"
 ---
+This page covers the host-side OpenShell gateway that NemoClaw uses for OpenClaw, Hermes, and LangChain Deep Agents Code sandboxes.
+It does not describe the in-sandbox agent gateway used by OpenClaw and Hermes.
+Deep Agents Code has no in-sandbox agent gateway.
+
 At every point, exactly one component owns the OpenShell gateway lifecycle.
 
 Platform images sometimes supervise the gateway with their own service.
@@ -149,15 +153,15 @@ A completed gateway step does not bypass listener, supervisor, identity, health,
 
 ## Inspect the selected authority
 
-Run `nemoclaw status` to see the management mode and redacted owner identity selected by the last onboarding run.
+Run `$$nemoclaw status` to see the management mode and redacted owner identity selected by the last onboarding run.
 The JSON form includes the same secret-free fields under `gatewayAuthority`.
 
 ```bash
-nemoclaw status
-nemoclaw status --json
+$$nemoclaw status
+$$nemoclaw status --json
 ```
 
-`nemoclaw debug` also records `gatewayAuthority` in `onboard-session-summary.txt`.
+`$$nemoclaw debug` also records `gatewayAuthority` in `onboard-session-summary.txt`.
 The status and debug views omit the external state directory and never include credential values.
 
 ## Keep external lifecycle paths inert
@@ -174,7 +178,7 @@ When a valid checkpoint exists, they compare the current authority with that che
 The comparison uses the exact gateway name and port.
 If the authority changed, teardown stops before it scans listeners or changes gateway runtime resources.
 
-`nemoclaw stop` does not scan or signal the externally supervised gateway.
+`$$nemoclaw stop` does not scan or signal the externally supervised gateway.
 Final-sandbox cleanup can stop local dashboard forwards and remove the modern local gateway registration.
 It does not signal the gateway, use the legacy `gateway destroy` fallback, or remove its Docker volumes.
 Uninstall can delete the selected sandboxes, providers, and local registration.
@@ -198,5 +202,5 @@ For `systemd-user`, use the user manager:
 systemctl --user status openshell-gateway.service
 ```
 
-Bring the supervised gateway up, leave exactly one verified process holding the configured port, and rerun `nemoclaw onboard`.
+Bring the supervised gateway up, leave exactly one verified process holding the configured port, and rerun `$$nemoclaw onboard`.
 To hand the lifecycle back to NemoClaw, remove the platform gateway service and declare `nemoclaw-managed`, or remove the declaration.
diff --git a/docs/deployment/install-openclaw-plugins.mdx b/docs/deployment/install-openclaw-plugins.mdx
@@ -10,6 +10,7 @@ content:
   type: "how_to"
 skill:
   priority: 20
+agent-variants: ["openclaw"]
 ---
 
 <Warning>

diff --git a/docs/deployment/sandbox-hardening.mdx b/docs/deployment/sandbox-hardening.mdx
@@ -8,6 +8,7 @@ description-agent: "Includes the sandbox container image hardening reference, co
 keywords: ["nemoclaw sandbox hardening", "container security", "docker capabilities", "process limits"]
 content:
   type: "reference"
+agent-variants: ["openclaw"]
 ---
 The NemoClaw sandbox image applies several security measures to reduce the attack surface and limit damage from untrusted workloads.
 

diff --git a/docs/get-started/quickstart-hermes.mdx b/docs/get-started/quickstart-hermes.mdx
@@ -10,6 +10,7 @@ content:
   type: "get_started"
 skill:
   priority: 20
+agent-variants: ["hermes"]
 ---
 Create a sandboxed Hermes agent, then chat with it from the dashboard or terminal.
 The `nemohermes` command is the NemoClaw CLI with Hermes pre-selected.

diff --git a/docs/get-started/quickstart-langchain-deepagents-code.mdx b/docs/get-started/quickstart-langchain-deepagents-code.mdx
@@ -13,6 +13,7 @@ audience: "operators"
 status: published
 content:
   type: "get_started"
+agent-variants: ["deepagents"]
 ---
 Create a sandboxed LangChain Deep Agents Code agent, then run your first prompt.
 The `nemo-deepagents` command is an alias for `nemoclaw` with the `langchain-deepagents-code` agent pre-selected.

diff --git a/docs/get-started/quickstart.mdx b/docs/get-started/quickstart.mdx
@@ -10,6 +10,7 @@ content:
   type: "get_started"
 skill:
   priority: 10
+agent-variants: ["openclaw"]
 ---
 Create a sandboxed OpenClaw agent, then send it a first prompt.
 

diff --git a/docs/index.yml b/docs/index.yml
@@ -313,7 +313,7 @@ navigation:
                 path: _build/agent-variants/deployment/deploy-to-headless-server.openclaw.generated.mdx
                 slug: deploy-to-headless-server
               - page: "Gateway Lifecycle Authority"
-                path: deployment/gateway-lifecycle-authority.mdx
+                path: _build/agent-variants/deployment/gateway-lifecycle-authority.openclaw.generated.mdx
                 slug: gateway-lifecycle-authority
           - section: "Monitoring"
             slug: monitoring
@@ -628,6 +628,9 @@ navigation:
               - page: "Deploy to a Headless Server"
                 path: _build/agent-variants/deployment/deploy-to-headless-server.deepagents.generated.mdx
                 slug: deploy-to-headless-server
+              - page: "Gateway Lifecycle Authority"
+                path: _build/agent-variants/deployment/gateway-lifecycle-authority.deepagents.generated.mdx
+                slug: gateway-lifecycle-authority
           - section: "Monitoring"
             slug: monitoring
             collapsed: open-by-default
@@ -664,6 +667,9 @@ navigation:
               - page: "Platform Support"
                 path: reference/platform-support.mdx
                 slug: platform-support
+              - page: "Enterprise Readiness"
+                path: _build/agent-variants/reference/enterprise-readiness.deepagents.generated.mdx
+                slug: enterprise-readiness
               - page: "Architecture Details"
                 path: _build/agent-variants/reference/architecture.deepagents.generated.mdx
                 slug: architecture
@@ -995,6 +1001,9 @@ navigation:
               - page: "Deploy to a Headless Server"
                 path: _build/agent-variants/deployment/deploy-to-headless-server.hermes.generated.mdx
                 slug: deploy-to-headless-server
+              - page: "Gateway Lifecycle Authority"
+                path: _build/agent-variants/deployment/gateway-lifecycle-authority.hermes.generated.mdx
+                slug: gateway-lifecycle-authority
           - section: "Monitoring"
             slug: monitoring
             collapsed: open-by-default

diff --git a/docs/inference/configure-model-capabilities.mdx b/docs/inference/configure-model-capabilities.mdx
@@ -8,6 +8,7 @@ description-agent: "Configures OpenClaw reasoning mode, reasoning effort, and te
 keywords: ["nemoclaw reasoning model", "nemoclaw reasoning effort", "nemoclaw vision model"]
 content:
   type: "how_to"
+agent-variants: ["openclaw"]
 ---
 
 Declare model capabilities before onboarding so NemoClaw can bake them into the OpenClaw configuration.

diff --git a/docs/inference/declarative-agents-manifest.mdx b/docs/inference/declarative-agents-manifest.mdx
@@ -21,6 +21,7 @@ content:
 skill:
   priority: 30
 status: published
+agent-variants: ["openclaw"]
 ---
 
 NemoClaw can bake a multi-agent OpenClaw layout into a sandbox image from a single checked-in manifest.

diff --git a/docs/inference/set-up-ollama.mdx b/docs/inference/set-up-ollama.mdx
@@ -10,6 +10,7 @@ content:
   type: "how_to"
 skill:
   priority: 10
+agent-variants: ["openclaw", "hermes"]
 ---
 Use Ollama when you want the default local inference setup path.
 NemoClaw detects Ollama on the host and can install, start, or upgrade it on supported systems.

diff --git a/docs/inference/set-up-sub-agent.mdx b/docs/inference/set-up-sub-agent.mdx
@@ -10,6 +10,7 @@ content:
   type: "how_to"
 skill:
   priority: 30
+agent-variants: ["openclaw"]
 ---
 OpenClaw documents the sub-agent behavior, `sessions_spawn` tool, `agents.list` configuration, tool policy, nesting, and auth model in [Sub-Agents](https://docs.openclaw.ai/tools/subagents).
 Use that page as the source of truth for how OpenClaw sub-agents work.

diff --git a/docs/inference/use-hermes-provider.mdx b/docs/inference/use-hermes-provider.mdx
@@ -8,6 +8,7 @@ description-agent: "Sets up Hermes Provider for Hermes Agent. Use when choosing
 keywords: ["Hermes Provider", "NemoClaw Hermes inference", "Hermes models"]
 content:
   type: "how_to"
+agent-variants: ["hermes"]
 ---
 Hermes Provider routes Hermes Agent through the host OpenShell provider that NemoClaw registers during Hermes onboarding.
 This option is available only for Hermes Agent.

diff --git a/docs/manage-sandboxes/add-channels-after-onboarding.mdx b/docs/manage-sandboxes/add-channels-after-onboarding.mdx
@@ -8,6 +8,7 @@ description-agent: "Explains channels list and channels add behavior, policy app
 keywords: ["nemoclaw channels add", "add messaging channel", "channels list"]
 content:
   type: "how_to"
+agent-variants: ["openclaw", "hermes"]
 ---
 Run channel commands from the host, not from inside the sandbox.
 

diff --git a/docs/manage-sandboxes/enable-channels-during-onboarding.mdx b/docs/manage-sandboxes/enable-channels-during-onboarding.mdx
@@ -8,6 +8,7 @@ description-agent: "Explains the interactive and scripted onboarding flows for s
 keywords: ["nemoclaw onboard messaging", "messaging channel picker", "channel environment variables"]
 content:
   type: "how_to"
+agent-variants: ["openclaw", "hermes"]
 ---
 Enable channels during onboarding when you are creating or recreating a sandbox.
 

diff --git a/docs/manage-sandboxes/gateway-lifecycle-control.mdx b/docs/manage-sandboxes/gateway-lifecycle-control.mdx
@@ -10,6 +10,7 @@ content:
   type: "concept"
 skill:
   priority: 20
+agent-variants: ["openclaw", "hermes"]
 ---
 Built-in OpenClaw and Hermes images support two direct-container lifecycle topologies for `recover` and `gateway restart`.
 

diff --git a/docs/manage-sandboxes/install-plugins-hermes.mdx b/docs/manage-sandboxes/install-plugins-hermes.mdx
@@ -10,6 +10,7 @@ content:
   type: "how_to"
 skill:
   priority: 40
+agent-variants: ["hermes"]
 ---
 Hermes plugins extend the Hermes runtime inside a NemoClaw-managed sandbox.
 They are different from NemoClaw skills and from OpenClaw plugins, so install them through the Hermes plugin path instead of `skill install`.

diff --git a/docs/manage-sandboxes/manage-messaging-channels.mdx b/docs/manage-sandboxes/manage-messaging-channels.mdx
@@ -8,6 +8,7 @@ description-agent: "Explains channel credential rotation, destructive removal, p
 keywords: ["nemoclaw channels remove", "nemoclaw channels stop", "messaging credential rotation", "channel conflicts"]
 content:
   type: "how_to"
+agent-variants: ["openclaw", "hermes"]
 ---
 Use host-side channel commands to change a configured messaging channel.
 

diff --git a/docs/manage-sandboxes/messaging-channels.mdx b/docs/manage-sandboxes/messaging-channels.mdx
@@ -10,6 +10,7 @@ content:
   type: "concept"
 skill:
   priority: 30
+agent-variants: ["openclaw", "hermes"]
 ---
 NemoClaw supports Telegram, Discord, Slack, WeChat, WhatsApp, and Microsoft Teams for OpenClaw and Hermes sandboxes.
 OpenShell-managed processes and gateway resources carry channel traffic.

diff --git a/docs/manage-sandboxes/runtime-controls.mdx b/docs/manage-sandboxes/runtime-controls.mdx
@@ -10,6 +10,7 @@ content:
   type: "concept"
 skill:
   priority: 10
+agent-variants: ["openclaw", "hermes"]
 ---
 Use this matrix to choose the operation that makes a sandbox change take effect.
 NemoClaw applies its security posture in three layers: what onboarding writes into the sandbox image, what the running sandbox can hot-reload, and what requires a rebuild or re-onboard.

diff --git a/docs/manage-sandboxes/set-up-discord.mdx b/docs/manage-sandboxes/set-up-discord.mdx
@@ -8,6 +8,7 @@ description-agent: "Explains Discord bot credentials and NemoClaw server, user,
 keywords: ["nemoclaw discord", "discord bot token", "discord server id"]
 content:
   type: "how_to"
+agent-variants: ["openclaw", "hermes"]
 ---
 Create a Discord bot and choose its server and user access before you enable the channel.
 

diff --git a/docs/manage-sandboxes/set-up-microsoft-teams.mdx b/docs/manage-sandboxes/set-up-microsoft-teams.mdx
@@ -8,6 +8,7 @@ description-agent: "Explains the experimental Microsoft Teams Bot Framework webh
 keywords: ["nemoclaw teams", "microsoft teams bot framework", "teams webhook"]
 content:
   type: "how_to"
+agent-variants: ["openclaw", "hermes"]
 ---
 Microsoft Teams support is experimental and uses a public Bot Framework webhook that forwards to the sandbox.
 

diff --git a/docs/manage-sandboxes/set-up-slack.mdx b/docs/manage-sandboxes/set-up-slack.mdx
@@ -8,6 +8,7 @@ description-agent: "Explains Slack bot and app token validation, allowlists, men
 keywords: ["nemoclaw slack", "slack socket mode", "slack app token", "slack allowlist"]
 content:
   type: "how_to"
+agent-variants: ["openclaw", "hermes"]
 ---
 Slack uses Socket Mode and requires both a bot token and an app-level token.
 

diff --git a/docs/manage-sandboxes/set-up-telegram.mdx b/docs/manage-sandboxes/set-up-telegram.mdx
@@ -8,6 +8,7 @@ description-agent: "Explains Telegram bot creation and NemoClaw allowlist, group
 keywords: ["nemoclaw telegram", "telegram bot token", "telegram allowed ids"]
 content:
   type: "how_to"
+agent-variants: ["openclaw", "hermes"]
 ---
 Create a Telegram bot and choose the direct-message and group behavior before you enable the channel.
 

diff --git a/docs/manage-sandboxes/set-up-wechat.mdx b/docs/manage-sandboxes/set-up-wechat.mdx
@@ -8,6 +8,7 @@ description-agent: "Explains the experimental personal WeChat iLink QR flow, pro
 keywords: ["nemoclaw wechat", "wechat qr", "wechat ilink"]
 content:
   type: "how_to"
+agent-variants: ["openclaw", "hermes"]
 ---
 WeChat support is experimental and uses Tencent's iLink gateway.
 The supported mode in this release is personal WeChat with `bot_type=3`.

diff --git a/docs/manage-sandboxes/set-up-whatsapp.mdx b/docs/manage-sandboxes/set-up-whatsapp.mdx
@@ -8,6 +8,7 @@ description-agent: "Explains the experimental in-sandbox WhatsApp QR pairing flo
 keywords: ["nemoclaw whatsapp", "whatsapp qr pairing", "whatsapp session"]
 content:
   type: "how_to"
+agent-variants: ["openclaw", "hermes"]
 ---
 WhatsApp support is experimental and pairs inside the sandbox rather than through a host-side token or OpenShell credential provider.
 

diff --git a/docs/monitoring/manage-deepagents-trace-export.mdx b/docs/monitoring/manage-deepagents-trace-export.mdx
@@ -8,6 +8,7 @@ description-agent: "Manages Deep Agents trace export. Use when stopping the coll
 keywords: ["manage deep agents traces", "disable nemoclaw observability", "remove otel collector"]
 content:
   type: "how_to"
+agent-variants: ["deepagents"]
 ---
 Manage sandbox instrumentation and the operator-owned host collector as separate lifecycle controls.
 NemoClaw does not start, stop, upgrade, or remove the collector.

diff --git a/docs/monitoring/monitor-sandbox-activity.mdx b/docs/monitoring/monitor-sandbox-activity.mdx
@@ -10,6 +10,7 @@ content:
   type: "how_to"
 skill:
   priority: 10
+agent-variants: ["openclaw", "hermes"]
 ---
 Use NemoClaw status commands, log streams, and OpenShell TUI views to inspect sandbox health, trace agent behavior, and diagnose problems.