docs: prepare v0.0.78 release notes by cjagwani · Pull Request #6529 · NVIDIA/NemoClaw

cjagwani · 2026-07-09T01:22:10Z

Summary

Prepares the user documentation for NemoClaw v0.0.78 by replacing the unreleased section with release highlights and synchronizing the affected inference, lifecycle, messaging, and CLI reference pages with merged behavior.

Changes

Publish the v0.0.78 release-notes section with links to the most specific user guides for each shipped behavior.
Document authoritative Deep Agents route health, Nemotron Ultra profile behavior, and Hermes compatible-endpoint context metadata.
Document forced rebuild recovery after total backup failure and the ownership-safe tunnel/full-stop behavior.
Keep command examples and shared agent variants aligned with the current OpenClaw, Hermes, and Deep Agents interfaces.

Source mapping:

#3787 -> docs/about/release-notes.mdx: Record reliable workspace template seeding during sandbox startup.
#4960 -> docs/about/release-notes.mdx: Record safer detection of rewritten OpenClaw gateway processes.
#5676 -> docs/about/release-notes.mdx: Record warning-tolerant agent-list JSON handling.
#5857 -> docs/about/release-notes.mdx: Record synchronization of explicit OpenClaw main-agent model state.
#5929 -> docs/about/release-notes.mdx: Record copyable SSH port-forward guidance for remote dashboards.
#6068 -> docs/about/release-notes.mdx: Record custom-image plugin provenance reconciliation.
#6116 -> docs/about/release-notes.mdx: Record live-loopback dashboard-forward recovery.
#6122 -> docs/about/release-notes.mdx: Announce validated, round-trippable policy YAML output.
#6211 -> docs/manage-sandboxes/lifecycle.mdx, docs/reference/commands.mdx, docs/about/release-notes.mdx: Explain the explicit no-backup rebuild --force recovery boundary.
#6283 -> docs/about/release-notes.mdx: Record Hermes WebUI port alignment.
#6293 -> docs/inference/switch-inference-providers.mdx, docs/about/release-notes.mdx: Document compatible-endpoint context-length probing for Hermes.
#6320 -> docs/about/release-notes.mdx: Record bounded gateway-recovery waits.
#6377 -> docs/reference/commands.mdx, docs/about/release-notes.mdx: Explain rebuild diagnostics and prepared MCP-destroy recovery.
#6412 -> docs/get-started/quickstart-langchain-deepagents-code.mdx, docs/about/release-notes.mdx: Document authoritative agent-visible inference route health.
#6421 -> docs/about/release-notes.mdx: Record the longer quiet-pull window for managed vLLM images.
#6431 -> docs/inference/model-capability-audit.mdx, docs/about/release-notes.mdx: Document the version-pinned Nemotron Ultra profile plugin.
#6439 -> docs/about/release-notes.mdx: Summarize the authenticated, pinned credential-capture helper boundary.
#6450 -> docs/manage-sandboxes/messaging-channels.mdx, docs/reference/commands.mdx, docs/about/release-notes.mdx: Document host-forward cleanup and ownership-safe gateway-port release.
#6474 -> docs/manage-sandboxes/messaging-channels.mdx, docs/about/release-notes.mdx: Record composable OpenClaw messaging runtime loaders.
#6475 -> docs/about/release-notes.mdx: Record removal of the unavailable Kimi K2.6 production endpoint option.
#6480 -> docs/about/release-notes.mdx: Record stderr routing for the plugin registration banner.
#6481 -> docs/about/release-notes.mdx: Record post-pull Ollama model discovery checks.
#6482 -> docs/about/release-notes.mdx: Record Ollama model warm-up after daemon restart.
#6486 -> docs/about/release-notes.mdx: Publish the opt-in, thread-scoped Deep Agents auto-approval boundary.
#6490 -> docs/about/release-notes.mdx: Record diagnostics for custom images missing the managed runtime.
#6494 -> docs/inference/model-capability-audit.mdx, docs/about/release-notes.mdx: Document nonempty tool-call content preservation and placeholder rejection.
#6497 -> docs/get-started/quickstart-langchain-deepagents-code.mdx, docs/about/release-notes.mdx: Document isolated Deep Agents route-probe output.
#6506 -> docs/get-started/quickstart-langchain-deepagents-code.mdx, docs/about/release-notes.mdx: Document observability-preserving managed route probes.
#6508 -> docs/about/release-notes.mdx: Link the new extension taxonomy and SDK-readiness reference from the release summary.

Release-source verification: GitHub reports all 29 cited source PRs as merged with base main, and every merge commit is an ancestor of origin/main at 17bf9a6a9688b3b1d69cf4b37d3f23110acb055e. No source-mapping mismatches were found.

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: Documentation-only release-prep changes; npm run docs validates variants, routes, and Fern content.
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:
Non-success, skipped, or missing CI check accepted by maintainer — check name, approval link, and follow-up issue:

Verification

PR description includes the DCO sign-off declaration and every commit appears as Verified in GitHub
Normal pre-commit, commit-msg, and pre-push hooks passed, or npm run check:diff passed when hooks were skipped or unavailable
Targeted behavior tests pass for the current change set, or tests are marked not applicable above — command/result or justification: Tests are not applicable to this documentation-only change set.
Applicable broad gate passed — npm test for broad runtime/test-harness changes; npm run check for repo-wide validation/coverage changes — command/result:
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) — exited 0 with zero errors; Fern reported the existing unauthenticated redirect-check and light-mode contrast warnings.
Doc pages follow the style guide (doc changes only)
New doc pages include SPDX header and frontmatter (new pages only)

Signed-off-by: Charan Jagwani cjagwani@nvidia.com

Signed-off-by: cjagwani <cjagwani@nvidia.com>

coderabbitai · 2026-07-09T01:22:42Z

📝 Walkthrough

Walkthrough

This PR updates NemoClaw documentation only: release notes now describe v0.0.78 changes, model capability audit rows switch to the Ultra model, model-metadata tuning docs split OpenClaw/Hermes behavior, and rebuild/stop command docs clarify backup-failure and gateway-ownership semantics.

Changes

Documentation updates

Layer / File(s)	Summary
Release notes and quickstart troubleshooting `docs/about/release-notes.mdx`, `docs/get-started/quickstart-langchain-deepagents-code.mdx`	Replaces "Unreleased" with a detailed v0.0.78 changelog and adds a troubleshooting note about rebuilding sandboxes missing a route-probe helper.
Model capability audit rows `docs/inference/model-capability-audit.mdx`	Interactive and headless dcode audit rows updated to the Ultra model with revised affordance expectations and source links.
OpenClaw/Hermes model metadata tuning `docs/inference/switch-inference-providers.mdx`	Splits "Tune Model Metadata" into OpenClaw and Hermes sections, revises `NEMOCLAW_CONTEXT_WINDOW` behavior, and clarifies unsupported agent-timeout variables for Hermes.
Rebuild failure/backup behavior `docs/manage-sandboxes/lifecycle.mdx`, `docs/reference/commands.mdx`	Documents partial-archive preservation, stop-before-delete on total backup failure, and the `--force` recovery path.
Tunnel/stop and messaging channel semantics `docs/manage-sandboxes/messaging-channels.mdx`, `docs/reference/commands.mdx`	Clarifies `channels stop` vs `tunnel stop` vs deprecated full stop, gateway ownership, and host-forward preservation rules.

Estimated code review effort: 2 (Simple) | ~10 minutes

Suggested reviewers: prekshivyas, jyaunches

🚥 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 is clear and relevant, and it accurately reflects the release-prep focus even though the PR also updates several related docs.

✨ 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/docs-v0-0-78-release-prep

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

github-actions · 2026-07-09T01:22:51Z

🌿 Preview your docs: https://nvidia-preview-pr-6529.docs.buildwithfern.com/nemoclaw

github-actions · 2026-07-09T01:22:57Z

E2E Advisor Recommendation

Required E2E: None
Optional E2E: None

Workflow run

Full advisor summary

E2E Recommendation Advisor

Base: origin/main
Head: HEAD
Confidence: high

Required E2E

None. No E2E is recommended because the diff is documentation-only. Although the docs describe runtime-sensitive areas such as onboarding, inference, rebuild, messaging, and stop/tunnel behavior, no executable code, policies, workflows, sandbox assets, or user-flow tests are changed.

Optional E2E

None.

New E2E recommendations

None.

github-actions · 2026-07-09T01:22:58Z

E2E Target Recommendation

Required E2E targets: None
Optional E2E targets: None

Workflow run

Full E2E target advisor summary

E2E Target Advisor

Base: origin/main
Head: HEAD
Confidence: high

Required E2E targets

None. Docs-only changes do not affect the E2E target registry, workflow machinery, live support wiring, fixtures, or runtime behavior, so no E2E target dispatch is required.

Optional E2E targets

None.

Relevant changed files

None.

github-actions · 2026-07-09T01:26:36Z

PR Review Advisor (Nemotron Ultra) — No blocking findings

Merge posture: No blocking advisor findings
Primary next action: Fix PRA-1: Release notes claim v0.0.78 features but PR only updates docs — verify code PRs merged before tag; then add or justify PRA-T1.
Open items: 1 required · 0 warnings · 1 suggestion · 5 test follow-ups
Since last review: 3 prior items resolved · 2 still apply · 0 new items found

Action checklist

PRA-1 Fix: Release notes claim v0.0.78 features but PR only updates docs — verify code PRs merged before tag in docs/about/release-notes.mdx:1
PRA-T1 Add or justify test follow-up: Acceptance clause
PRA-T2 Add or justify test follow-up: Acceptance clause
PRA-T3 Add or justify test follow-up: Acceptance clause
PRA-T4 Add or justify test follow-up: Acceptance clause
PRA-T5 Add or justify test follow-up: Acceptance clause
PRA-2 In-scope improvement: Deep Agents Ultra audit rows document placeholder rejection but remain 'not-yet-run' in docs/inference/model-capability-audit.mdx:107

Findings index

ID	Severity	Category	Location	Required action
`PRA-1`	Required	acceptance	`docs/about/release-notes.mdx:1`	Maintainer must verify each release note bullet maps to a merged code PR (e.g., #6486, #6412, #6293, #6122, #6211, #6068, #5929, #6439) before cutting the v0.0.78 release tag.
`PRA-2`	Improvement	acceptance	`docs/inference/model-capability-audit.mdx:107`	Add audit evidence row when Ultra profile validation completes; should include trajectory showing placeholder rejection before shell dispatch. Or add a note that this is planned but not yet validated.

🚨 Required before merge

Address these before merging unless a maintainer explicitly overrides the advisor with rationale.

`PRA-1` Required — Release notes claim v0.0.78 features but PR only updates docs — verify code PRs merged before tag

Location: docs/about/release-notes.mdx:1
Category: acceptance
Problem: Release notes describe 8 major v0.0.78 features (thread-scoped auto-approval feat(dcode): add thread-scoped auto-approval #6486, authoritative inference health fix(status): make inference route health authoritative #6412, max_model_len probing fix(inference): honor compatible-endpoint max_model_len for Hermes context (#6177) #6293, policy-get round-trip feat(policy): add sandbox policy get command for clean YAML output #6122, rebuild --force recovery fix(sandbox): allow rebuild --force to skip backup when container is unreachable #6211, custom image plugin provenance fix(sandbox): harden PR 6065 advisor follow-ups #6068, remote dashboard SSH hints fix(cli): show SSH port-forward hint for remote dashboard access (#5925) #5929, credential form hardening fix(security): harden local credential capture #6439) but this PR only updates documentation. No code diff in this PR to verify these features exist in the v0.0.78 codebase.
Impact: If any cited feature PR is not merged on main, release notes will be inaccurate and users will see documented features that don't work.
Required action: Maintainer must verify each release note bullet maps to a merged code PR (e.g., feat(dcode): add thread-scoped auto-approval #6486, fix(status): make inference route health authoritative #6412, fix(inference): honor compatible-endpoint max_model_len for Hermes context (#6177) #6293, feat(policy): add sandbox policy get command for clean YAML output #6122, fix(sandbox): allow rebuild --force to skip backup when container is unreachable #6211, fix(sandbox): harden PR 6065 advisor follow-ups #6068, fix(cli): show SSH port-forward hint for remote dashboard access (#5925) #5929, fix(security): harden local credential capture #6439) before cutting the v0.0.78 release tag.
Expected follow-up: Fix before merge or get explicit maintainer override.
Verification: Check git log on main for each cited PR: feat(dcode): add thread-scoped auto-approval feat(dcode): add thread-scoped auto-approval #6486, fix(status): make inference route health authoritative fix(status): make inference route health authoritative #6412, fix(inference): honor compatible-endpoint max_model_len fix(inference): honor compatible-endpoint max_model_len for Hermes context (#6177) #6293, fix(policy): add sandbox policy get feat(policy): add sandbox policy get command for clean YAML output #6122, fix(rebuild): --force recovery fix(sandbox): allow rebuild --force to skip backup when container is unreachable #6211, fix(installer): custom image plugin provenance fix(sandbox): harden PR 6065 advisor follow-ups #6068, fix(remote): SSH port-forward hints fix(cli): show SSH port-forward hint for remote dashboard access (#5925) #5929, fix(security): harden local credential capture fix(security): harden local credential capture #6439
Missing regression test: N/A — documentation accuracy finding; no test needed. Maintainers should verify each release note bullet maps to a merged code PR before cutting the release tag.
Done when: The required change is committed and verification passes: Check git log on main for each cited PR: feat(dcode): add thread-scoped auto-approval feat(dcode): add thread-scoped auto-approval #6486, fix(status): make inference route health authoritative fix(status): make inference route health authoritative #6412, fix(inference): honor compatible-endpoint max_model_len fix(inference): honor compatible-endpoint max_model_len for Hermes context (#6177) #6293, fix(policy): add sandbox policy get feat(policy): add sandbox policy get command for clean YAML output #6122, fix(rebuild): --force recovery fix(sandbox): allow rebuild --force to skip backup when container is unreachable #6211, fix(installer): custom image plugin provenance fix(sandbox): harden PR 6065 advisor follow-ups #6068, fix(remote): SSH port-forward hints fix(cli): show SSH port-forward hint for remote dashboard access (#5925) #5929, fix(security): harden local credential capture fix(security): harden local credential capture #6439.
Evidence: PR diff shows only .mdx file changes; no code changes. PR description lists 30+ source PRs mapping to doc changes.

Review findings by urgency: 1 required fix, 0 items to resolve/justify, 1 in-scope improvement

⚠️ Resolve or justify before merge

Investigate these in the current review; either fix them, explain why they are not applicable, or document the accepted risk.

None.

💡 In-scope improvements

These are lower-risk, not throwaway. Prefer fixing them in this PR when they are local to changed code; defer only with rationale or a linked follow-up.

`PRA-2` Improvement — Deep Agents Ultra audit rows document placeholder rejection but remain 'not-yet-run'

Location: docs/inference/model-capability-audit.mdx:107
Category: acceptance
Problem: Audit rows updated from Nemotron 3 Super to Nemotron 3 Ultra with new 'Required affordance' text about Ultra profile plugin preserving nonempty tool-call content and rejecting '[content]' execute placeholder before shell dispatch. This documents a new security-relevant model behavior but the audit rows remain 'not-yet-run' with no evidence.
Impact: Audit matrix claims a security feature (placeholder rejection before shell dispatch) but has no validation evidence. Users may assume it's tested when it's only documented.
Suggested action: Add audit evidence row when Ultra profile validation completes; should include trajectory showing placeholder rejection before shell dispatch. Or add a note that this is planned but not yet validated.
Expected follow-up: Prefer a current-PR fix when local to changed code; defer only with rationale or linked follow-up.
Verification: Verify agents/langchain-deepagents-code/profile-plugin/ exists and implements the placeholder rejection described. Check generate-config.ts loads this plugin for Ultra.
Missing regression test: Add audit evidence row when Ultra profile validation completes; should include trajectory showing placeholder rejection before shell dispatch.
Done when: The local improvement is applied, or the PR notes why it should be deferred.
Evidence: model-capability-audit.mdx lines 107-116: Deep Agents rows now reference Ultra profile plugin with 'rejects the observed literal [content] execute placeholder before shell dispatch' but State='not-yet-run'

Test follow-ups to resolve or justify

If these cover changed behavior, prefer adding them in this PR; otherwise state why existing coverage is enough or link the follow-up.

PRA-T1 Acceptance clause — Thread-scoped auto-approval for managed Deep Agents Code (feat(dcode): add thread-scoped auto-approval #6486) — add test evidence or identify existing coverage. Release notes bullet references feat(dcode): add thread-scoped auto-approval #6486; maintainer must verify PR merged on main before tag
PRA-T2 Acceptance clause — Authoritative agent-visible inference health (fix(status): make inference route health authoritative #6412) — add test evidence or identify existing coverage. Release notes bullet references fix(status): make inference route health authoritative #6412; maintainer must verify PR merged on main before tag
PRA-T3 Acceptance clause — Compatible-endpoint max_model_len probing for Hermes (fix(inference): honor compatible-endpoint max_model_len for Hermes context (#6177) #6293) — add test evidence or identify existing coverage. Release notes bullet references fix(inference): honor compatible-endpoint max_model_len for Hermes context (#6177) #6293; maintainer must verify PR merged on main before tag
PRA-T4 Acceptance clause — Round-trippable policy export via policy-get (feat(policy): add sandbox policy get command for clean YAML output #6122) — add test evidence or identify existing coverage. Release notes bullet references feat(policy): add sandbox policy get command for clean YAML output #6122; maintainer must verify PR merged on main before tag
PRA-T5 Acceptance clause — Rebuild --force recovery after total backup failure (fix(sandbox): allow rebuild --force to skip backup when container is unreachable #6211) — add test evidence or identify existing coverage. Release notes bullet references fix(sandbox): allow rebuild --force to skip backup when container is unreachable #6211; maintainer must verify PR merged on main before tag

Since last review details

Current findings, using the urgency labels above:

`PRA-1` Required — Release notes claim v0.0.78 features but PR only updates docs — verify code PRs merged before tag

Location: docs/about/release-notes.mdx:1
Category: acceptance
Problem: Release notes describe 8 major v0.0.78 features (thread-scoped auto-approval feat(dcode): add thread-scoped auto-approval #6486, authoritative inference health fix(status): make inference route health authoritative #6412, max_model_len probing fix(inference): honor compatible-endpoint max_model_len for Hermes context (#6177) #6293, policy-get round-trip feat(policy): add sandbox policy get command for clean YAML output #6122, rebuild --force recovery fix(sandbox): allow rebuild --force to skip backup when container is unreachable #6211, custom image plugin provenance fix(sandbox): harden PR 6065 advisor follow-ups #6068, remote dashboard SSH hints fix(cli): show SSH port-forward hint for remote dashboard access (#5925) #5929, credential form hardening fix(security): harden local credential capture #6439) but this PR only updates documentation. No code diff in this PR to verify these features exist in the v0.0.78 codebase.
Impact: If any cited feature PR is not merged on main, release notes will be inaccurate and users will see documented features that don't work.
Required action: Maintainer must verify each release note bullet maps to a merged code PR (e.g., feat(dcode): add thread-scoped auto-approval #6486, fix(status): make inference route health authoritative #6412, fix(inference): honor compatible-endpoint max_model_len for Hermes context (#6177) #6293, feat(policy): add sandbox policy get command for clean YAML output #6122, fix(sandbox): allow rebuild --force to skip backup when container is unreachable #6211, fix(sandbox): harden PR 6065 advisor follow-ups #6068, fix(cli): show SSH port-forward hint for remote dashboard access (#5925) #5929, fix(security): harden local credential capture #6439) before cutting the v0.0.78 release tag.
Expected follow-up: Fix before merge or get explicit maintainer override.
Verification: Check git log on main for each cited PR: feat(dcode): add thread-scoped auto-approval feat(dcode): add thread-scoped auto-approval #6486, fix(status): make inference route health authoritative fix(status): make inference route health authoritative #6412, fix(inference): honor compatible-endpoint max_model_len fix(inference): honor compatible-endpoint max_model_len for Hermes context (#6177) #6293, fix(policy): add sandbox policy get feat(policy): add sandbox policy get command for clean YAML output #6122, fix(rebuild): --force recovery fix(sandbox): allow rebuild --force to skip backup when container is unreachable #6211, fix(installer): custom image plugin provenance fix(sandbox): harden PR 6065 advisor follow-ups #6068, fix(remote): SSH port-forward hints fix(cli): show SSH port-forward hint for remote dashboard access (#5925) #5929, fix(security): harden local credential capture fix(security): harden local credential capture #6439
Missing regression test: N/A — documentation accuracy finding; no test needed. Maintainers should verify each release note bullet maps to a merged code PR before cutting the release tag.
Done when: The required change is committed and verification passes: Check git log on main for each cited PR: feat(dcode): add thread-scoped auto-approval feat(dcode): add thread-scoped auto-approval #6486, fix(status): make inference route health authoritative fix(status): make inference route health authoritative #6412, fix(inference): honor compatible-endpoint max_model_len fix(inference): honor compatible-endpoint max_model_len for Hermes context (#6177) #6293, fix(policy): add sandbox policy get feat(policy): add sandbox policy get command for clean YAML output #6122, fix(rebuild): --force recovery fix(sandbox): allow rebuild --force to skip backup when container is unreachable #6211, fix(installer): custom image plugin provenance fix(sandbox): harden PR 6065 advisor follow-ups #6068, fix(remote): SSH port-forward hints fix(cli): show SSH port-forward hint for remote dashboard access (#5925) #5929, fix(security): harden local credential capture fix(security): harden local credential capture #6439.
Evidence: PR diff shows only .mdx file changes; no code changes. PR description lists 30+ source PRs mapping to doc changes.

`PRA-2` Improvement — Deep Agents Ultra audit rows document placeholder rejection but remain 'not-yet-run'

Location: docs/inference/model-capability-audit.mdx:107
Category: acceptance
Problem: Audit rows updated from Nemotron 3 Super to Nemotron 3 Ultra with new 'Required affordance' text about Ultra profile plugin preserving nonempty tool-call content and rejecting '[content]' execute placeholder before shell dispatch. This documents a new security-relevant model behavior but the audit rows remain 'not-yet-run' with no evidence.
Impact: Audit matrix claims a security feature (placeholder rejection before shell dispatch) but has no validation evidence. Users may assume it's tested when it's only documented.
Suggested action: Add audit evidence row when Ultra profile validation completes; should include trajectory showing placeholder rejection before shell dispatch. Or add a note that this is planned but not yet validated.
Expected follow-up: Prefer a current-PR fix when local to changed code; defer only with rationale or linked follow-up.
Verification: Verify agents/langchain-deepagents-code/profile-plugin/ exists and implements the placeholder rejection described. Check generate-config.ts loads this plugin for Ultra.
Missing regression test: Add audit evidence row when Ultra profile validation completes; should include trajectory showing placeholder rejection before shell dispatch.
Done when: The local improvement is applied, or the PR notes why it should be deferred.
Evidence: model-capability-audit.mdx lines 107-116: Deep Agents rows now reference Ultra profile plugin with 'rejects the observed literal [content] execute placeholder before shell dispatch' but State='not-yet-run'

Workflow run details

This is an automated, non-binding review; it still expects maintainers and agents to respond to each required or warning item. Treat suggestions as current-PR improvements when they touch changed code; defer only with maintainer rationale or a linked follow-up. A human maintainer must make the final merge decision.

github-actions · 2026-07-09T01:27:53Z

PR Review Advisor — No blocking findings

Merge posture: No blocking advisor findings
Primary next action: Add or justify PRA-T1 and any related test follow-ups.
Open items: 0 required · 0 warnings · 0 suggestions · 1 test follow-up
Since last review: 1 prior item resolved · 0 still apply · 0 new items found

Action checklist

PRA-T1 Add or justify test follow-up: Acceptance clause

Test follow-ups to resolve or justify

If these cover changed behavior, prefer adding them in this PR; otherwise state why existing coverage is enough or link the follow-up.

PRA-T1 Acceptance clause — No linked issue clauses were provided in the deterministic linkedIssues context for this PR. — add test evidence or identify existing coverage. The validation context reported linkedIssues: []; therefore there were no issue-body or issue-comment acceptance clauses to map literally.

Workflow run details

This is an automated, non-binding review; it still expects maintainers and agents to respond to each required or warning item. Treat suggestions as current-PR improvements when they touch changed code; defer only with maintainer rationale or a linked follow-up. A human maintainer must make the final merge decision.

coderabbitai

Actionable comments posted: 1

🧹 Nitpick comments (1)

docs/reference/commands.mdx (1)
2636-2649: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick win

Consider deduplicating the $$nemoclaw stop ownership-rule description.

The gateway-port/host-forward ownership semantics for the deprecated $$nemoclaw stop command are described twice — briefly under tunnel stop (Lines 2646-2649) and in full under the dedicated $$nemoclaw stop section (Lines 2677-2681) — with slightly different wording each time (e.g., Line 2679's host-forward cleanup detail isn't mentioned in the brief version). Two independently-maintained descriptions of the same nuanced behavior risk diverging on future edits.

Consider trimming the brief mention to a pointer to the dedicated section instead of restating the rules.
✏️ Proposed simplification
 `$$nemoclaw stop` remains as a deprecated legacy full stop.
-In addition to stopping tunnel services, it attempts to stop the selected agent's host forwards when the sandbox uses a manifest-resolved non-OpenClaw agent.
-It also attempts to safely release an unshared OpenShell gateway port whose ownership NemoClaw can verify.
-Shared gateways remain running, and ambiguous ownership fails closed without releasing the port.
+Refer to [`$$nemoclaw stop`](#$$nemoclaw-stop) for its host-forward and gateway-port ownership rules.
Also applies to: 2670-2681
🤖 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/commands.mdx` around lines 2636 - 2649, The ownership-rule
explanation for the deprecated $$nemoclaw stop command is duplicated in both the
tunnel stop description and the dedicated $$nemoclaw stop section, which can
drift over time. Trim the brief mention in the tunnel stop docs so it only
points readers to the full $$nemoclaw stop section, and keep the complete
ownership semantics in one place using the existing $$nemoclaw stop and tunnel
stop references.

🤖 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/switch-inference-providers.mdx`:
- Around line 338-350: The onboarding docs currently imply
`model.context_length` is always written, but it should only be emitted when
`NEMOCLAW_CONTEXT_WINDOW` resolves to a valid value. Update the wording in the
`switch-inference-providers.mdx` section that describes Hermes,
`NEMOCLAW_CONTEXT_WINDOW`, and the generated `config.yaml` so it clearly states
that valid overrides populate `model.context_length`, while unset or invalid
values leave the field absent.

---

Nitpick comments:
In `@docs/reference/commands.mdx`:
- Around line 2636-2649: The ownership-rule explanation for the deprecated
$$nemoclaw stop command is duplicated in both the tunnel stop description and
the dedicated $$nemoclaw stop section, which can drift over time. Trim the brief
mention in the tunnel stop docs so it only points readers to the full $$nemoclaw
stop section, and keep the complete ownership semantics in one place using the
existing $$nemoclaw stop and tunnel stop references.

🪄 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: 45a4015c-4aa2-4e6e-86a8-b44b61d3a1d4

📥 Commits

Reviewing files that changed from the base of the PR and between 17bf9a6 and 71dbd2c.

📒 Files selected for processing (7)

docs/about/release-notes.mdx
docs/get-started/quickstart-langchain-deepagents-code.mdx
docs/inference/model-capability-audit.mdx
docs/inference/switch-inference-providers.mdx
docs/manage-sandboxes/lifecycle.mdx
docs/manage-sandboxes/messaging-channels.mdx
docs/reference/commands.mdx

cjagwani · 2026-07-09T01:31:42Z

Addressed the automated review findings in f80b12d2b521ebc2f2c8485f8353b46f69bb0a69.

Release-note source verification: all 29 PRs cited in the PR body report MERGED with base main, and every merge commit is an ancestor of origin/main at 17bf9a6a9688b3b1d69cf4b37d3f23110acb055e. No source-mapping mismatches were found.
Hermes metadata wording: clarified that model.context_length is written after either a valid explicit override or a successful endpoint probe, and remains absent only when neither path yields a valid value.
Hermes rebuild notes: removed unsupported OpenClaw-only max-token, reasoning, and heartbeat settings from the Hermes variant.
Deep Agents helper version: verified that dcode-managed-exec first landed in fix(connect): preserve DCode observability during probes #6506 after v0.0.77, then documented the v0.0.78 client and pre-v0.0.78 image boundary.
Model audit state: no evidence row was added. The page already defines not-yet-run as no completed evidence, says source-seeded rows are not live benchmark claims, and lists the evidence still required for placeholder rejection.
Hermes implementation verification: the Hermes Dockerfile, build environment parser, and config generator consume only NEMOCLAW_CONTEXT_WINDOW for generated model metadata. Relevant filtered tests passed 9/9.
Rebuild implementation verification: --yes/-y only skips confirmation; --force also authorizes continuing after total backup failure. Relevant filtered tests passed 3/3.

Post-fix verification: npm run docs exited 0 with zero errors, and normal commit/push hooks passed. Fern continues to report the existing unauthenticated redirect-check and light-mode contrast warnings described in the PR body.

cjagwani · 2026-07-09T01:36:18Z

Existing source-PR coverage for the advisor test follow-ups:

PRA-T1 / feat(dcode): add thread-scoped auto-approval #6486 thread-scoped auto-approval: test/langchain-deepagents-code-direct-module-patch.test.ts covers opt-in activation and thread-boundary reset; test/langchain-deepagents-code-managed-entrypoints.test.ts covers explicit activation only in thread-opt-in mode; src/lib/onboard/dcode-auto-approval.test.ts covers the agent/capability gate. A reusable live acceptance check also exists at test/e2e/e2e-cloud-experimental/checks/12-deepagents-code-thread-auto-approval.sh.
PRA-T2 / fix(status): make inference route health authoritative #6412 authoritative inference health: src/lib/actions/sandbox/status-inference.test.ts, src/lib/actions/sandbox/inference-route-health.test.ts, src/lib/actions/sandbox/doctor-inference.test.ts, and test/cli/sandbox-status-json.test.ts cover broken, healthy, and missing-trust inference.local route results.
PRA-T3 / fix(inference): honor compatible-endpoint max_model_len for Hermes context (#6177) #6293 Hermes context probing: test/generate-hermes-config.test.ts covers the endpoint probe through generated model.context_length; test/compatible-endpoint-context-probe.test.ts covers propagation from a real loopback /v1/models response.
PRA-T4 / feat(policy): add sandbox policy get command for clean YAML output #6122 policy export: src/lib/actions/sandbox/policy-get.test.ts covers metadata stripping into round-trippable YAML; src/commands/sandbox/policy/get.test.ts covers CLI output; test/policy-roundtrip-docs.test.ts covers the documented export/reapply flow.
PRA-T5 / fix(sandbox): allow rebuild --force to skip backup when container is unreachable #6211 forced rebuild recovery: src/lib/actions/sandbox/rebuild-flow-helpers.test.ts, rebuild-backup-phase.test.ts, and rebuild-restore-phase.test.ts cover force continuation, non-force aborts, recorded skipped-backup state, and the final warning. All 47 tests in those three files pass locally.

These tests landed with the cited source PRs and their commits are ancestors of current main. PR #6529 is documentation-only, has no linked issue acceptance clauses, and does not need duplicate behavior tests.

## Summary Prepares the user documentation for NemoClaw v0.0.78 by replacing the unreleased section with release highlights and synchronizing the affected inference, lifecycle, messaging, and CLI reference pages with merged behavior. ## Changes - Publish the v0.0.78 release-notes section with links to the most specific user guides for each shipped behavior. - Document authoritative Deep Agents route health, Nemotron Ultra profile behavior, and Hermes compatible-endpoint context metadata. - Document forced rebuild recovery after total backup failure and the ownership-safe tunnel/full-stop behavior. - Keep command examples and shared agent variants aligned with the current OpenClaw, Hermes, and Deep Agents interfaces. Source mapping: - [NVIDIA#3787](NVIDIA#3787) -> `docs/about/release-notes.mdx`: Record reliable workspace template seeding during sandbox startup. - [NVIDIA#4960](NVIDIA#4960) -> `docs/about/release-notes.mdx`: Record safer detection of rewritten OpenClaw gateway processes. - [NVIDIA#5676](NVIDIA#5676) -> `docs/about/release-notes.mdx`: Record warning-tolerant agent-list JSON handling. - [NVIDIA#5857](NVIDIA#5857) -> `docs/about/release-notes.mdx`: Record synchronization of explicit OpenClaw main-agent model state. - [NVIDIA#5929](NVIDIA#5929) -> `docs/about/release-notes.mdx`: Record copyable SSH port-forward guidance for remote dashboards. - [NVIDIA#6068](NVIDIA#6068) -> `docs/about/release-notes.mdx`: Record custom-image plugin provenance reconciliation. - [NVIDIA#6116](NVIDIA#6116) -> `docs/about/release-notes.mdx`: Record live-loopback dashboard-forward recovery. - [NVIDIA#6122](NVIDIA#6122) -> `docs/about/release-notes.mdx`: Announce validated, round-trippable policy YAML output. - [NVIDIA#6211](NVIDIA#6211) -> `docs/manage-sandboxes/lifecycle.mdx`, `docs/reference/commands.mdx`, `docs/about/release-notes.mdx`: Explain the explicit no-backup `rebuild --force` recovery boundary. - [NVIDIA#6283](NVIDIA#6283) -> `docs/about/release-notes.mdx`: Record Hermes WebUI port alignment. - [NVIDIA#6293](NVIDIA#6293) -> `docs/inference/switch-inference-providers.mdx`, `docs/about/release-notes.mdx`: Document compatible-endpoint context-length probing for Hermes. - [NVIDIA#6320](NVIDIA#6320) -> `docs/about/release-notes.mdx`: Record bounded gateway-recovery waits. - [NVIDIA#6377](NVIDIA#6377) -> `docs/reference/commands.mdx`, `docs/about/release-notes.mdx`: Explain rebuild diagnostics and prepared MCP-destroy recovery. - [NVIDIA#6412](NVIDIA#6412) -> `docs/get-started/quickstart-langchain-deepagents-code.mdx`, `docs/about/release-notes.mdx`: Document authoritative agent-visible inference route health. - [NVIDIA#6421](NVIDIA#6421) -> `docs/about/release-notes.mdx`: Record the longer quiet-pull window for managed vLLM images. - [NVIDIA#6431](NVIDIA#6431) -> `docs/inference/model-capability-audit.mdx`, `docs/about/release-notes.mdx`: Document the version-pinned Nemotron Ultra profile plugin. - [NVIDIA#6439](NVIDIA#6439) -> `docs/about/release-notes.mdx`: Summarize the authenticated, pinned credential-capture helper boundary. - [NVIDIA#6450](NVIDIA#6450) -> `docs/manage-sandboxes/messaging-channels.mdx`, `docs/reference/commands.mdx`, `docs/about/release-notes.mdx`: Document host-forward cleanup and ownership-safe gateway-port release. - [NVIDIA#6474](NVIDIA#6474) -> `docs/manage-sandboxes/messaging-channels.mdx`, `docs/about/release-notes.mdx`: Record composable OpenClaw messaging runtime loaders. - [NVIDIA#6475](NVIDIA#6475) -> `docs/about/release-notes.mdx`: Record removal of the unavailable Kimi K2.6 production endpoint option. - [NVIDIA#6480](NVIDIA#6480) -> `docs/about/release-notes.mdx`: Record stderr routing for the plugin registration banner. - [NVIDIA#6481](NVIDIA#6481) -> `docs/about/release-notes.mdx`: Record post-pull Ollama model discovery checks. - [NVIDIA#6482](NVIDIA#6482) -> `docs/about/release-notes.mdx`: Record Ollama model warm-up after daemon restart. - [NVIDIA#6486](NVIDIA#6486) -> `docs/about/release-notes.mdx`: Publish the opt-in, thread-scoped Deep Agents auto-approval boundary. - [NVIDIA#6490](NVIDIA#6490) -> `docs/about/release-notes.mdx`: Record diagnostics for custom images missing the managed runtime. - [NVIDIA#6494](NVIDIA#6494) -> `docs/inference/model-capability-audit.mdx`, `docs/about/release-notes.mdx`: Document nonempty tool-call content preservation and placeholder rejection. - [NVIDIA#6497](NVIDIA#6497) -> `docs/get-started/quickstart-langchain-deepagents-code.mdx`, `docs/about/release-notes.mdx`: Document isolated Deep Agents route-probe output. - [NVIDIA#6506](NVIDIA#6506) -> `docs/get-started/quickstart-langchain-deepagents-code.mdx`, `docs/about/release-notes.mdx`: Document observability-preserving managed route probes. - [NVIDIA#6508](NVIDIA#6508) -> `docs/about/release-notes.mdx`: Link the new extension taxonomy and SDK-readiness reference from the release summary. Release-source verification: GitHub reports all 29 cited source PRs as merged with base `main`, and every merge commit is an ancestor of `origin/main` at `17bf9a6a9688b3b1d69cf4b37d3f23110acb055e`. No source-mapping mismatches were found. ## Type of Change - [ ] Code change (feature, bug fix, or refactor) - [ ] Code change with doc updates - [ ] Doc only (prose changes, no code sample modifications) - [x] Doc only (includes code sample changes) ## Quality Gates  - [ ] Tests added or updated for changed behavior - [ ] Existing tests cover changed behavior — justification: - [x] Tests not applicable — justification: Documentation-only release-prep changes; `npm run docs` validates variants, routes, and Fern content. - [x] 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: - [ ] Non-success, skipped, or missing CI check accepted by maintainer — check name, approval link, and follow-up issue: ## Verification  - [x] PR description includes the DCO sign-off declaration and every commit appears as `Verified` in GitHub - [x] Normal `pre-commit`, `commit-msg`, and `pre-push` hooks passed, or `npm run check:diff` passed when hooks were skipped or unavailable - [x] Targeted behavior tests pass for the current change set, or tests are marked not applicable above — command/result or justification: Tests are not applicable to this documentation-only change set. - [ ] Applicable broad gate passed — `npm test` for broad runtime/test-harness changes; `npm run check` for repo-wide validation/coverage changes — command/result: - [x] Quality Gates section completed with required justifications or waivers - [x] No secrets, API keys, or credentials committed - [ ] `npm run docs` builds without warnings (doc changes only) — exited 0 with zero errors; Fern reported the existing unauthenticated redirect-check and light-mode contrast warnings. - [x] Doc pages follow the [style guide](https://github.com/NVIDIA/NemoClaw/blob/main/docs/CONTRIBUTING.md) (doc changes only) - [ ] New doc pages include SPDX header and frontmatter (new pages only) ---  Signed-off-by: Charan Jagwani <cjagwani@nvidia.com> --------- Signed-off-by: cjagwani <cjagwani@nvidia.com>

docs: prepare v0.0.78 release notes

71dbd2c

Signed-off-by: cjagwani <cjagwani@nvidia.com>

cjagwani added area: docs Documentation, examples, guides, or docs build v0.0.78 Release target labels Jul 9, 2026

cjagwani self-assigned this Jul 9, 2026

cjagwani requested a review from cv July 9, 2026 01:23

Merge branch 'main' into codex/docs-v0-0-78-release-prep

309241d

cv approved these changes Jul 9, 2026

View reviewed changes

coderabbitai Bot reviewed Jul 9, 2026

View reviewed changes

Comment thread docs/inference/switch-inference-providers.mdx Outdated

docs: clarify v0.0.78 metadata boundaries

f80b12d

github-actions Bot mentioned this pull request Jul 9, 2026

docs(get-started): streamline quickstarts #6527

Merged

21 tasks

cv merged commit ce19c7b into main Jul 9, 2026
45 checks passed

cv deleted the codex/docs-v0-0-78-release-prep branch July 9, 2026 01:37

Uh oh!

Conversation

cjagwani commented Jul 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Changes

Type of Change

Quality Gates

Verification

Uh oh!

coderabbitai Bot commented Jul 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

Uh oh!

github-actions Bot commented Jul 9, 2026

Uh oh!

github-actions Bot commented Jul 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

E2E Advisor Recommendation

E2E Recommendation Advisor

Required E2E

Optional E2E

New E2E recommendations

Uh oh!

github-actions Bot commented Jul 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

E2E Target Recommendation

E2E Target Advisor

Required E2E targets

Optional E2E targets

Relevant changed files

Uh oh!

github-actions Bot commented Jul 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR Review Advisor (Nemotron Ultra) — No blocking findings

Action checklist

Findings index

🚨 Required before merge

PRA-1 Required — Release notes claim v0.0.78 features but PR only updates docs — verify code PRs merged before tag

⚠️ Resolve or justify before merge

💡 In-scope improvements

PRA-2 Improvement — Deep Agents Ultra audit rows document placeholder rejection but remain 'not-yet-run'

PRA-1 Required — Release notes claim v0.0.78 features but PR only updates docs — verify code PRs merged before tag

PRA-2 Improvement — Deep Agents Ultra audit rows document placeholder rejection but remain 'not-yet-run'

Uh oh!

github-actions Bot commented Jul 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR Review Advisor — No blocking findings

Action checklist

Uh oh!

coderabbitai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

cjagwani commented Jul 9, 2026

Uh oh!

cjagwani commented Jul 9, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

cjagwani commented Jul 9, 2026 •

edited

Loading

coderabbitai Bot commented Jul 9, 2026 •

edited

Loading

github-actions Bot commented Jul 9, 2026 •

edited

Loading

github-actions Bot commented Jul 9, 2026 •

edited

Loading

github-actions Bot commented Jul 9, 2026 •

edited

Loading

`PRA-1` Required — Release notes claim v0.0.78 features but PR only updates docs — verify code PRs merged before tag

`PRA-2` Improvement — Deep Agents Ultra audit rows document placeholder rejection but remain 'not-yet-run'

`PRA-1` Required — Release notes claim v0.0.78 features but PR only updates docs — verify code PRs merged before tag

`PRA-2` Improvement — Deep Agents Ultra audit rows document placeholder rejection but remain 'not-yet-run'

github-actions Bot commented Jul 9, 2026 •

edited

Loading