docs: single-source starter prompt in Markdown

.github/workflows/docs-preview-pr.yaml

@@ -13,6 +13,8 @@ on:
       - "README.md"
       - "package.json"
       - "package-lock.json"
+      - "scripts/generate-starter-prompt.ts"
+      - "scripts/sync-agent-variant-docs.ts"
       - ".github/workflows/docs-preview-pr.yaml"
 
 concurrency:
@@ -56,8 +58,10 @@ jobs:
       - name: Install docs dependencies
         run: npm ci --ignore-scripts
 
-      - name: Validate docs
-        run: npm run docs
+      - name: Prepare and validate docs
+        run: |
+          npm run docs:prepare
+          npm run docs:validate
 
       - name: Generate preview URL
         if: ${{ steps.fern-preview.outputs.enabled == 'true' }}
@@ -66,7 +70,6 @@ jobs:
           FERN_TOKEN: ${{ secrets.FERN_TOKEN }}
           PREVIEW_ID: pr-${{ github.event.pull_request.number }}
         run: |
-          npm run docs:check-agent-variants
           FERN_VERSION=$(node -p "require('./fern/fern.config.json').version")
           set +e
           OUTPUT=$(cd fern && npx --yes "fern-api@${FERN_VERSION}" generate --docs --instance "$FERN_STAGING_INSTANCE" --preview --id "$PREVIEW_ID" 2>&1)

.github/workflows/docs-publish-public.yaml

@@ -47,10 +47,10 @@ jobs:
       - name: Install docs dependencies
         run: npm ci --ignore-scripts
 
-      - name: Generate doc variants and validate Fern docs
+      - name: Prepare and validate Fern docs
         run: |
-          npm run docs:sync-agent-variants
-          npm run docs
+          npm run docs:prepare
+          npm run docs:validate
 
       - name: Publish public docs
         env:

.github/workflows/docs-publish-staging.yaml

@@ -11,6 +11,7 @@ on:
       - "fern/**"
       - "package.json"
       - "package-lock.json"
+      - "scripts/generate-starter-prompt.ts"
       - "scripts/sync-agent-variant-docs.ts"
       - ".github/workflows/docs-publish-staging.yaml"
 
@@ -43,10 +44,10 @@ jobs:
       - name: Install docs dependencies
         run: npm ci --ignore-scripts
 
-      - name: Generate doc variants and validate Fern docs
+      - name: Prepare and validate Fern docs
         run: |
-          npm run docs:sync-agent-variants
-          npm run docs
+          npm run docs:prepare
+          npm run docs:validate
 
       - name: Publish staging docs
         env:

.pre-commit-config.yaml

@@ -171,7 +171,7 @@ repos:
         name: Repository checks
         entry: npm run checks
         language: system
-        files: ^(\.pre-commit-config\.yaml$|\.github/workflows/e2e\.yaml$|Dockerfile(?:\.base)?$|agents/openclaw/manifest\.yaml$|agents/hermes/(?:Dockerfile(?:\.base)?|manifest\.yaml|mcp-config-transaction\.py)$|nemoclaw-blueprint/blueprint\.yaml$|nemoclaw/package\.json$|scripts/(?:brev-launchable-ci-cpu|check-installer-hash|install-openshell|update-hermes-agent)\.sh$|src/lib/actions/sandbox/openshell-child-visible-credentials\.v[0-9]+\.[0-9]+\.[0-9]+\.json$|bin/.*\.(cjs|js|mjs)$|src/.*\.(cts|mts|ts|tsx)$|scripts/.*\.(cjs|cts|js|mjs|mts|ts|tsx)$|test/.*\.(cjs|cts|js|mjs|mts|ts|tsx)$|nemoclaw/src/.*\.(cts|mts|ts|tsx)$)
+        files: ^(\.pre-commit-config\.yaml$|\.github/workflows/e2e\.yaml$|Dockerfile(?:\.base)?$|agents/openclaw/manifest\.yaml$|agents/hermes/(?:Dockerfile(?:\.base)?|manifest\.yaml|mcp-config-transaction\.py)$|docs/resources/starter-prompt\.md$|nemoclaw-blueprint/blueprint\.yaml$|nemoclaw/package\.json$|scripts/(?:brev-launchable-ci-cpu|check-installer-hash|install-openshell|update-hermes-agent)\.sh$|src/lib/actions/sandbox/openshell-child-visible-credentials\.v[0-9]+\.[0-9]+\.[0-9]+\.json$|bin/.*\.(cjs|js|mjs)$|src/.*\.(cts|mts|ts|tsx)$|scripts/.*\.(cjs|cts|js|mjs|mts|ts|tsx)$|test/.*\.(cjs|cts|js|mjs|mts|ts|tsx)$|nemoclaw/src/.*\.(cts|mts|ts|tsx)$)
         pass_filenames: false
         priority: 10
 

ci/source-shape-test-budget.json

@@ -456,6 +456,11 @@
       "test": "walkthrough.sh does not embed NVIDIA_INFERENCE_API_KEY in tmux or sandbox commands",
       "category": "security"
     },
+    {
+      "file": "test/starter-prompt-docs.test.ts",
+      "test": "prepares the Starter Prompt in every docs build entry point (#5048)",
+      "category": "compatibility"
+    },
     {
       "file": "test/validate-blueprint.test.ts",
       "test": "pins the sandbox image by digest instead of a mutable tag (#1438)",

docs/CONTRIBUTING.md

@@ -109,6 +109,33 @@ Public docs publish automatically when a `v*.*.*` release tag is pushed.
 The public publish job runs in the `docs-public` environment, verifies that the tag commit is reachable from `origin/main`, regenerates agent variants, validates Fern docs, and publishes to the public Fern instance.
 If the tag does not point to a commit on `main`, the job stops before installing dependencies or running Fern.
 
+## Starter Prompt Generation
+
+The canonical coding-agent installation prompt lives in `docs/resources/starter-prompt.md`.
+Edit that Markdown file instead of placing prompt text in a React component.
+Downstream consumers can pin the source with a raw URL such as
+`https://raw.githubusercontent.com/NVIDIA/NemoClaw/<commit-sha>/docs/resources/starter-prompt.md`.
+The Markdown SPDX comment is part of that raw file but does not appear when Markdown is rendered.
+
+The `scripts/generate-starter-prompt.ts` script removes the Markdown SPDX preamble and writes `docs/_build/StarterPrompt.generated.mdx`.
+The generated snippet wraps the prompt in Fern's native visible `Prompt` component, which displays the prompt body and supplies the copy button.
+The generated file is ignored by Git and is recreated by the docs build.
+
+Run the generator directly when you need to inspect the generated snippet:
+
+```bash
+npm run docs:sync-starter-prompt
+```
+
+Run the read-only comparison after generation when you need to verify that the snippet matches the Markdown source:
+
+```bash
+npm run docs:check-starter-prompt
+```
+
+The shared `npm run docs:prepare` step generates the Starter Prompt and agent variants.
+The normal `npm run docs`, `npm run docs:live`, agent-variant sync, preview-watcher, and docs publish workflows run that step before Fern validates, serves, previews, or publishes the pages that include the prompt.
+
 ## Agent Variant Generation
 
 Some Fern pages appear in the OpenClaw, Hermes, and Deep Agents guide variants.

docs/about/overview.mdx

@@ -95,7 +95,7 @@ Use these topics to learn more about NemoClaw and how to install and use it.
 - Read [Architecture Overview](how-it-works) to understand how NemoClaw works.
 - Read [Ecosystem](ecosystem) to understand how your agent, OpenShell, and NemoClaw relate in the wider stack, and when to use NemoClaw versus OpenShell.
 - Follow [Quickstart with OpenClaw](../get-started/quickstart) to install NemoClaw and run your first OpenClaw sandbox.
-- Read [AI Agent Docs](../resources/agent-skills) to let your AI coding assistant fetch NemoClaw Markdown docs.
+- Read [Use NemoClaw Docs with Your Coding Agents](../resources/agent-skills) to let your AI coding assistant fetch NemoClaw Markdown docs.
 - Review [Community Solutions](../resources/community-contributions) for community-driven blueprint examples, showcases, and integrations.
 - Read [Choose an Inference Provider](../inference/learn-and-choose/choose-inference-provider) to check the inference providers that NemoClaw supports and how inference routing works.
 
@@ -105,7 +105,7 @@ Use these topics to learn more about NemoClaw and how to install and use it.
 - Read [Architecture Overview](how-it-works) to understand how NemoClaw works.
 - Read [Ecosystem](ecosystem) to understand how Hermes, OpenShell, and NemoClaw relate in the wider stack, and when to use NemoClaw versus OpenShell.
 - Follow [Quickstart with Hermes](../get-started/quickstart) to install NemoClaw and run your first Hermes sandbox with `$$nemoclaw`.
-- Read [AI Agent Docs](../resources/agent-skills) to let your AI coding assistant fetch NemoClaw Markdown docs.
+- Read [Use NemoClaw Docs with Your Coding Agents](../resources/agent-skills) to let your AI coding assistant fetch NemoClaw Markdown docs.
 - Review [Community Solutions](../resources/community-contributions) for community-driven blueprint examples, showcases, and integrations.
 - Read [Choose an Inference Provider](../inference/learn-and-choose/choose-inference-provider) to check the inference providers that NemoClaw supports and how inference routing works.
 
@@ -115,7 +115,7 @@ Use these topics to learn more about NemoClaw and how to install and use it.
 - Read [Architecture Overview](how-it-works) to understand how NemoClaw works.
 - Read [Ecosystem](ecosystem) to understand how Deep Agents, OpenShell, and NemoClaw relate in the wider stack, and when to use NemoClaw versus OpenShell.
 - Follow [Quickstart with Deep Agents](../get-started/quickstart) to install NemoClaw and run your first Deep Agents sandbox with `$$nemoclaw`.
-- Read [AI Agent Docs](../resources/agent-skills) to let your AI coding assistant fetch NemoClaw Markdown docs.
+- Read [Use NemoClaw Docs with Your Coding Agents](../resources/agent-skills) to let your AI coding assistant fetch NemoClaw Markdown docs.
 - Review [Community Solutions](../resources/community-contributions) for community-driven blueprint examples, showcases, and integrations.
 - Read [Choose an Inference Provider](../inference/learn-and-choose/choose-inference-provider) to check the inference providers that NemoClaw supports and how inference routing works.
 

docs/about/release-notes.mdx

@@ -201,7 +201,7 @@ NemoClaw v0.0.78 adds opt-in thread-scoped auto-approval and policy-routed repos
   For more information, refer to [Deploy to a Remote GPU Host](/user-guide/openclaw/deployment/deploy-to-remote-gpu), [NemoClaw Quickstart with Hermes](/user-guide/hermes/get-started/quickstart), [Choose Messaging Channels](/user-guide/openclaw/manage-sandboxes/messaging-channels/choose-messaging-channels), and [NemoClaw CLI Commands Reference](../reference/commands).
 - Starter prompts now bind the local credential form to an authenticated one-shot helper with immutable commit and SHA-256 pins, explicit isolated or account-home execution profiles, denial of ambient process-control variables, and a preview, edit, and confirm flow.
   This keeps credential collection behind verified helper and approved-command boundaries.
-  For more information, refer to [NemoClaw Quickstart with OpenClaw](/user-guide/openclaw/get-started/quickstart) and [Use NemoClaw Agent Prompts, Docs MCP Server, and Skills with Your AI Coding Agent](../resources/agent-skills).
+  For more information, refer to [NemoClaw Quickstart with OpenClaw](/user-guide/openclaw/get-started/quickstart) and [Use NemoClaw Docs with Your Coding Agents](../resources/agent-skills).
 
 ## v0.0.77
 
@@ -218,7 +218,7 @@ NemoClaw v0.0.77 hardens LangChain Deep Agents Code packaging, tracing, and guid
   For more information, refer to [Quickstart with LangChain Deep Agents Code](/user-guide/deepagents/get-started/quickstart), [Credential Storage](../security/credential-storage), and [Security Best Practices](../security/best-practices).
 - Starter prompts now reuse a checked-in local credential form with loopback-only submission, a restrictive content security policy, redacted confirmation output, and no external resources.
   This gives coding agents one reviewed credential-capture path instead of asking them to generate credential forms or collect secrets in chat.
-  For more information, refer to [NemoClaw Quickstart with OpenClaw](../get-started/quickstart) and [Use NemoClaw Agent Prompts, Docs MCP Server, and Skills with Your AI Coding Agent](../resources/agent-skills).
+  For more information, refer to [NemoClaw Quickstart with OpenClaw](../get-started/quickstart) and [Use NemoClaw Docs with Your Coding Agents](../resources/agent-skills).
 
 ## v0.0.76
 
@@ -450,7 +450,7 @@ NemoClaw v0.0.67 improves onboarding recovery, messaging reliability, OpenClaw a
   For more information, refer to [Choose Messaging Channels](/user-guide/openclaw/manage-sandboxes/messaging-channels/choose-messaging-channels).
 - AI-agent documentation now uses one compact routing skill instead of generated page-copy skills.
   The `nemoclaw-user-guide` skill points agents to the docs MCP server, `llms.txt`, and canonical Markdown routes so the docs site remains the source of truth.
-  For more information, refer to [Use NemoClaw Agent Prompts, Docs MCP Server, and Skills with Your AI Coding Agent](../resources/agent-skills).
+  For more information, refer to [Use NemoClaw Docs with Your Coding Agents](../resources/agent-skills).
 
 ## v0.0.66
 

docs/get-started/prerequisites.mdx

@@ -101,4 +101,4 @@ The table comes from [`ci/platform-matrix.json`](https://github.com/NVIDIA/NemoC
 
 - Use [Prepare Windows for NemoClaw](prerequisites/windows-preparation) if you are using Windows.
 - [Quickstart](quickstart) installs NemoClaw and launches your first sandboxed agent.
-- [AI Agent Docs](../resources/agent-skills) lets your AI coding assistant fetch NemoClaw Markdown docs before setup.
+- [Use NemoClaw Docs with Your Coding Agents](../resources/agent-skills) lets your AI coding assistant fetch NemoClaw Markdown docs before setup.

docs/get-started/quickstart-hermes.mdx

@@ -11,25 +11,20 @@ content:
 skill:
   priority: 20
 ---
-import { StarterPromptFallback } from "../_components/StarterPrompt";
-import { StarterPromptButton } from "../_components/StarterPromptButton";
-
 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.
 
-## Set Up with a Coding Agent
+## Set Up with the Starter Prompt on Your Coding Agent
 
 Copy this starter prompt into Cursor, Claude Code, Codex, Copilot, or another local coding agent when you want it to guide the installation.
-The prompt points the agent to [AI Agent Docs](../resources/agent-skills), this quickstart, the Markdown docs, and the optional `nemoclaw-user-guide` skill.
+The prompt points the agent to [Use NemoClaw Docs with Your Coding Agents](../resources/agent-skills), this quickstart, the Markdown docs, and the optional `nemoclaw-user-guide` skill.
 It asks the agent to confirm Hermes before it runs commands that create a sandbox or receive credentials and to use the checked-in local credential helper and form only after you approve the exact command that receives credentials.
 
-<StarterPromptButton />
-
-<StarterPromptFallback />
+<Markdown src="/../docs/_build/StarterPrompt.generated.mdx" />
 
-If you prefer to control setup directly, use the interactive installer in [Create Your First Sandboxed Agent](#create-your-first-sandboxed-agent).
+If you prefer to control setup directly, use [Set Up with the Interactive Installer on Your Terminal](#set-up-with-the-interactive-installer-on-your-terminal).
 
-## Create Your First Sandboxed Agent
+## Set Up with the Interactive Installer on Your Terminal
 
 If you use the coding-agent prompt in the preceding section, you can skip this procedure or keep it as reference.
 The prompt directs your coding agent to this quickstart, so it has the full setup context.