docs: split post-tag documentation tasks

[miyoungc]

Summary

Splits three oversized post-release documentation sections into focused task pages while preserving existing routes and section anchors. The new ownership boundaries make runtime identity, dual-DGX Station vLLM, and OpenClaw memory search easier to find and maintain without changing product behavior.

Changes

• Move the OpenClaw runtime identity procedure from Architecture to a focused reference task and preserve the existing Architecture anchor.
• Move first-run dual-DGX Station qualification and managed vLLM operation to a shared focused page, with agent-aware shell-installer commands.
• Move OpenClaw memory-search setup from the Ollama page to Configure Agents and preserve the existing Ollama anchor.
• Update navigation, quickstarts, Station preparation, and the v0.0.97 changelog to use the canonical pages.
• Remove colons from the sandbox native-package review title and SEC headings while keeping that evidence page outside Fern navigation.
• Add focused ownership, route, command, and variant regression tests.

Type of Change

• Code change (feature, bug fix, or refactor)
• Code change with doc updates
• Doc only (prose changes, no code sample modifications)
• Doc only (includes code sample changes)

Quality Gates

• Tests added or updated for changed behavior
• Existing tests cover changed behavior — justification:
• Tests not applicable — justification:
• Docs updated for user-facing behavior changes
• Docs not applicable — justification:
• Sensitive paths changed (security, policy, credentials, preflight, onboarding, inference, runner, sandbox, or messaging)
• Sensitive-path review completed or maintainer-approved waiver recorded — reviewer/approval link/justification: Documentation-only inference and security paths; commands were traced to checked-in source, and the exact-head documentation writer review plus nine-category security review passed at 6835d54da.
• Non-success, skipped, or missing CI check accepted by maintainer — check name, approval link, and follow-up issue:

Documentation Writer Review

• Documentation writer subagent reviewed the completed changes
• Result: docs-updated
• Evidence: Exact-head review of the complete documentation diff and generated agent variants passed; focused validation passed 5 files and 72 tests; npm run docs passed with 0 errors and 2 existing warnings.
• Agent: Codex Desktop

DGX Station Hardware Evidence

• Tested on DGX Station
• Tested commit: Not applicable; no Station preparation script changed.
• Station profile/scenario: Not applicable.
• Result: Not applicable.
• Supporting evidence: Not applicable.

Verification

• PR description includes a Signed-off-by: line and every commit appears as Verified in GitHub
• Normal pre-commit, commit-msg, and pre-push hooks passed, or npm run validate:pr passed after refreshing origin/main when hooks were skipped or unavailable
• Targeted behavior tests pass for the current change set, or tests are marked not applicable above — npx vitest run test/inference-options-docs.test.ts test/station-doc-ownership.test.ts test/docs-refactor-ownership.test.ts test/check-docs-published-routes.test.ts test/check-docs-links.test.ts: 5 files and 72 tests passed.
• Applicable broad gate passed — npm test for broad runtime/test-harness changes; npm run check for repo-wide validation/coverage changes — command/result: Not applicable; this is a documentation-only ownership refactor with focused regression tests.
• Quality Gates section completed with required justifications or waivers
• No secrets, API keys, or credentials committed
• npm run docs builds without warnings (doc changes only)
• Doc pages follow the style guide (doc changes only)
• New doc pages include SPDX header and frontmatter (new pages only)

npm run docs completed with zero errors and the repository's two existing Fern warnings.

---

Signed-off-by: Miyoung Choi miyoungc@nvidia.com
Signed-off-by: Prekshi Vyas prekshiv@nvidia.com

Summary by CodeRabbit

• New Features

  • Added a dedicated guide for distributed vLLM setup on two DGX Stations, including paired workflow prerequisites and cleanup.
  • Added guidance for configuring Memory Search and experimental Runtime Identity (including OpenClaw-specific runtime identity setup).

• Documentation

  • Updated get-started and inference pages with clearer cross-links to the new distributed vLLM, Memory Search, and Runtime Identity guides.
  • Improved Memory Search “shields up/down” instructions with timeout/reason, plus added verification wording.
  • Refreshed architecture and changelog navigation/content for clarity.

• Tests

  • Expanded documentation ownership and navigation checks to validate required content, links, and dual- vs single-station expectations.

[copy-pr-bot]

Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

[coderabbitai]

📝 Walkthrough

Walkthrough

The documentation adds dedicated guides for two-Station vLLM serving, host-Ollama memory search, and experimental runtime identity. Navigation, setup references, changelog entries, architecture guidance, and documentation ownership tests are updated.

Changes

Two-station vLLM documentation

Layer / File(s)	Summary
Distributed vLLM workflow guide docs/inference/set-up-vllm-on-two-dgx-stations.mdx	Documents trusted-pair qualification, preparation, resume behavior, runtime topology, network restrictions, hardening, and cleanup ownership.
Single-host vLLM guidance and cross-references docs/inference/set-up-vllm.mdx, docs/get-started/*	Clarifies single-Station behavior and links the deferred distributed workflow from setup and preparation pages.
Navigation and ownership validation docs/index.yml, test/inference-options-docs.test.ts, test/station-doc-ownership.test.ts, docs/changelog/2026-07-28.mdx	Adds variant navigation and validates links, peer qualification, generated installer content, trust-boundary wording, and release-note references.

Memory search configuration

Layer / File(s)	Summary
Host Ollama memory-search workflow docs/configure-agents/configure-memory-search.mdx	Adds instructions for embedding-server preparation, bridge access, bounded shields changes, provider configuration, index rebuilding, and verification.
Memory-search routing and validation docs/inference/set-up-ollama.mdx, docs/index.yml, test/docs-refactor-ownership.test.ts, docs/changelog/2026-07-28.mdx	Redirects existing guidance, adds navigation, and verifies content placement.

Experimental runtime identity

Layer / File(s)	Summary
Runtime identity configuration workflow docs/reference/configure-runtime-identity.mdx	Documents Okta and Microsoft Entra profiles, trust boundaries, conformance evidence, validation, apply behavior, secret scoping, inspection, and rollback.
Architecture reference and navigation handoff docs/reference/architecture.mdx, docs/index.yml, docs/changelog/2026-07-28.mdx	Updates gateway architecture guidance, condenses runtime-identity guidance, and links the dedicated reference.
Ownership validation test/docs-refactor-ownership.test.ts	Verifies runtime-identity content remains on the dedicated reference page.

Editorial documentation alignment

Layer / File(s)	Summary
Sandbox review headings docs/security/sandbox-base-2026-07-27-native-package-review.md	Updates the document title and SEC-1/SEC-2 heading formatting.

Estimated code review effort: 4 (Complex) | ~45 minutes

Possibly related PRs

• NVIDIA/NemoClaw#7686: Adds overlapping Microsoft Entra runtime-identity and architecture documentation.

Suggested labels: chore

Suggested reviewers: cv, jyaunches, prekshivyas

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title matches the main change: splitting post-release documentation into focused task pages.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch codex/post-tag-docs-audit-20260729-100159

---

_{Comment @coderabbitai help to get the list of available commands.}

[github-code-quality]

Code Coverage Overview

Languages: TypeScript

TypeScript / code-coverage/plugin

The overall coverage in commit 6835d54 in the codex/post-tag-docs-... branch remains at 96%, unchanged from commit b1f67f7 in the main branch.

TypeScript / code-coverage/cli

The overall coverage in commit 6835d54 in the codex/post-tag-docs-... branch remains at 81%, unchanged from commit 25de83d in the main branch.

Show a code coverage summary of the most impacted files.

File	main 25de83d	codex/post-tag-docs-... 6835d54	+/-
src/lib/onboard...ndbox-create.ts	83%	33%	-50%
src/lib/onboard...-create-plan.ts	88%	75%	-13%
src/lib/onboard...ndbox-create.ts	91%	83%	-8%
src/lib/domain/.../connect-env.ts	97%	89%	-8%
src/lib/sandbox...rce-identity.ts	88%	88%	0%

---

_{Updated July 30, 2026 18:35 UTC}

[github-actions]

🌿 Preview your docs: https://nvidia-preview-pr-7874.docs.buildwithfern.com/nemoclaw

[github-actions]

PR Review Advisor — No blocking findings reported

Advisor assessment: No blocking advisor findings reported
Next action: No advisor follow-up needed.
Findings: 0 blockers · 0 warnings · 0 suggestions

Model lanes

• GPT-5.6 Terra (primary): Completed · high confidence · 0 blockers · 0 warnings · 0 suggestions
• Nemotron 3 Ultra (second opinion): Completed · high confidence · 0 blockers · 0 warnings · 0 suggestions
• Model comparison: normalized findings match; normalized E2E selections match; severity counts match.

Nemotron output stays in workflow artifacts and does not change the assessment above.

E2E guidance

Advisory only. E2E / PR Gate selects and runs jobs independently.

Recommended E2E: None

Workflow run details

This automated review informs maintainers. Warnings and suggestions do not require a response. A maintainer decides whether to merge.

[coderabbitai]

Actionable comments posted: 1

🧹 Nitpick comments (1)

docs/reference/configure-runtime-identity.mdx (1)

87-87: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick win

Clarify that client_id_env must be OKTA_CLIENT_ID.

The YAML example presents client_id_env as a configurable field, while Line 125 states the reviewed policy fixes the source to OKTA_CLIENT_ID. A short inline note avoids readers trying a tenant-specific name and hitting a validation failure.

📝 Suggested wording

-  client_id_env: OKTA_CLIENT_ID
+  client_id_env: OKTA_CLIENT_ID # must be exactly this value

Also applies to: 125-126

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@docs/reference/configure-runtime-identity.mdx` at line 87, Add an inline YAML
comment beside client_id_env in the configuration example, and reinforce the
same clarification near the policy statement around the referenced later
section: document that the value must remain exactly OKTA_CLIENT_ID and is not
tenant-configurable.

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@docs/inference/set-up-vllm.mdx`:
- Around line 157-160: Update the Docker precedence statement in the NemoClaw
setup documentation to say that DOCKER_HOST takes precedence over DOCKER_CONTEXT
when both are set, while leaving the surrounding socket and Hugging Face cache
behavior unchanged.

---

Nitpick comments:
In `@docs/reference/configure-runtime-identity.mdx`:
- Line 87: Add an inline YAML comment beside client_id_env in the configuration
example, and reinforce the same clarification near the policy statement around
the referenced later section: document that the value must remain exactly
OKTA_CLIENT_ID and is not tenant-configurable.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 25dbb20b-0964-47f4-88d7-cac0ab67477f

📥 Commits

Reviewing files that changed from the base of the PR and between 376beb5 and 0214a73.

📒 Files selected for processing (16)

• docs/changelog/2026-07-28.mdx
• docs/configure-agents/configure-memory-search.mdx
• docs/get-started/dgx-station-preparation.mdx
• docs/get-started/quickstart-hermes.mdx
• docs/get-started/quickstart-langchain-deepagents-code.mdx
• docs/get-started/quickstart.mdx
• docs/index.yml
• docs/inference/set-up-ollama.mdx
• docs/inference/set-up-vllm-on-two-dgx-stations.mdx
• docs/inference/set-up-vllm.mdx
• docs/reference/architecture.mdx
• docs/reference/configure-runtime-identity.mdx
• docs/security/sandbox-base-2026-07-27-native-package-review.md
• test/docs-refactor-ownership.test.ts
• test/inference-options-docs.test.ts
• test/station-doc-ownership.test.ts

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[coderabbitai]

Actionable comments posted: 2

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@docs/reference/configure-runtime-identity.mdx`:
- Line 52: Split each compound documentation line in the runtime identity
reference, including the line containing “TC-INF-13” and the corresponding lines
at the other referenced locations, so every sentence occupies its own line.
Ensure each sentence ends with a period while preserving the existing wording
and meaning.
- Around line 156-157: Resolve the documentation link-policy issue in the
permission guidance around the Microsoft Graph User.Read and offline_access
references: either remove or replace the Microsoft Learn links with approved
in-repository guidance, unless the repository explicitly permits official
Microsoft Entra documentation links. Preserve the instructions about delegated
permissions, refresh-token issuance, and refresh-token replacement.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 6968bc94-cb96-451c-820f-a326cbf94e7e

📥 Commits

Reviewing files that changed from the base of the PR and between 0214a73 and 40c0f86.

📒 Files selected for processing (2)

• docs/reference/configure-runtime-identity.mdx
• test/inference-options-docs.test.ts

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@test/station-doc-ownership.test.ts`:
- Line 124: Update the ownership assertion in station-doc-ownership.test.ts to
exclude the current public wording, “The two-Station path is a Deferred
evaluation,” instead of the obsolete “dual-Station” phrase. Keep the assertion
focused on stationPreparation and the observable documentation-ownership
boundary.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: af0c03a3-5d30-4be7-bffd-bb78093058e2

📥 Commits

Reviewing files that changed from the base of the PR and between 40c0f86 and 776b344.

📒 Files selected for processing (7)

• docs/configure-agents/configure-memory-search.mdx
• docs/get-started/dgx-station-preparation.mdx
• docs/inference/set-up-vllm-on-two-dgx-stations.mdx
• docs/reference/architecture.mdx
• docs/reference/configure-runtime-identity.mdx
• test/docs-refactor-ownership.test.ts
• test/station-doc-ownership.test.ts

🚧 Files skipped from review as they are similar to previous changes (3)

• docs/get-started/dgx-station-preparation.mdx
• docs/reference/architecture.mdx
• docs/reference/configure-runtime-identity.mdx

[prekshivyas]

Reviewed exact head 6835d54. The documentation ownership split, security-sensitive procedures, generated variants, anchors, and regression tests are correct. Exact-head documentation and nine-category security reviews pass; focused tests and docs validation pass; no unresolved findings.