## Release range

- Previous release: `v0.0.128` at `ad6ba127fb9eb0e52436699f45933bb0850d906b`
- Candidate: `26922313bba96184e65c3663b351683ebae9504d`
- Candidate selection: historical
- Historical candidate exception: The maintainer selected the original release candidate to retain its reviewed documentation and accepted E2E evidence, excluding later product changes and newly observed failures.
- Commits: 16
- Risky files detected: 55

## QA context

### Risky areas

- Workflow / enforcement
- Installer / bootstrap
- Credentials / inference
- Onboarding / host glue
- Sandbox / policy / SSRF

### Suggested test focus

- Fresh install and upgrade paths
- Onboarding wizard and sandbox creation
- Policy enforcement, network egress, and SSRF protections
- CI checks, pre-commit hooks, and DCO declarations
- Credential storage and inference provider routing

## Canonical release entry

- Path: `docs/changelog/2026-09-23.mdx` at the candidate. The original release entry exists and remains included.
- Historical release-entry decision: retain this entry and exclude later release-note additions for changes outside the selected range.

## v0.0.129

NemoClaw v0.0.129 extends configuration export and deferred onboarding, improves OpenClaw restart readiness, and corrects recovery across local gateways.
It also strengthens host OS reporting and Windows WSL GPU verification.

- Configuration export preserves effective OpenClaw and Hermes settings when v1alpha1 omission defaults would change behavior.
  Supported OpenClaw exports for attached Ollama with a managed proxy and the fixed managed vLLM profile now include local inference service templates.
  These templates contain `image: null`; replace that value with an official, immutable, compatible v1 runtime image before v1 parsing or planning.
  Related changes: [PR #12054](https://github.com/NVIDIA/NemoClaw/pull/12054) and [PR #12161](https://github.com/NVIDIA/NemoClaw/pull/12161).
  For supported profiles and refusal conditions, refer to the [NemoClaw CLI Commands Reference](/user-guide/openclaw/reference/commands).
- Hermes and LangChain Deep Agents Code can defer onboarding when installing without NVIDIA hosted inference credentials.
  The installer creates no provider, sandbox, or completed onboarding state in that case; present credentials retain normal validation.
  Complete setup later through the installed agent's onboarding command.
  Related change: [PR #10341](https://github.com/NVIDIA/NemoClaw/pull/10341).
  For the opt-in flag and prerequisites, refer to the [CLI Commands Reference](/user-guide/deepagents/reference/commands).
- OpenClaw gateway restart loads updated native plugins from a fresh module graph and waits for `/readyz` to return HTTP `200` with `{"ready":true}` before repairing forwards.
  Managed startup also preserves the SQLite temporary-directory environment for standalone memory commands.
  Related changes: [PR #11911](https://github.com/NVIDIA/NemoClaw/pull/11911) and [PR #12256](https://github.com/NVIDIA/NemoClaw/pull/12256).
  Refer to [Install OpenClaw Plugins](/user-guide/openclaw/manage-sandboxes/install-openclaw-plugins) and [Understand Gateway Lifecycle Control](/user-guide/openclaw/manage-sandboxes/configure-sandboxes/understand-gateway-lifecycle-control).
- Sandbox status and recovery use the owning local gateway registry even when another gateway is active.
  Recovery guidance follows the recorded runtime provider, and rebuild delegates to the owning registry after confirmation.
  Resumed host-local inference onboarding reserves its retained authority before creating a runtime and reports actionable diagnostics when that reservation fails.
  Related changes: [PR #11945](https://github.com/NVIDIA/NemoClaw/pull/11945) and [PR #12075](https://github.com/NVIDIA/NemoClaw/pull/12075).
  Refer to [Recover and Rebuild Sandboxes](/user-guide/openclaw/manage-sandboxes/operate-sandboxes/recover-and-rebuild-sandboxes).
- Host readiness reports the Linux distribution and release, with warnings for unqualified or inconclusive OS evidence.
  Windows on Arm GPU verification checks every reported device and aggregate capacity without qualifying generic multi-GPU hosts for larger automatic Ollama model selection.
  Related changes: [PR #11780](https://github.com/NVIDIA/NemoClaw/pull/11780) and [PR #12212](https://github.com/NVIDIA/NemoClaw/pull/12212).
  Refer to [System Readiness](/user-guide/openclaw/reference/system-readiness) and [Prepare Windows for NemoClaw](/user-guide/openclaw/get-started/additional-setup/windows-preparation).
- NVIDIA Endpoints onboarding no longer selects the retired MiniMax M3 model.
  Interactive onboarding asks for a replacement; noninteractive onboarding uses an eligible configured model or the agent default.
  OpenRouter model selection remains separate from NVIDIA's retirement policy.
  Related change: [PR #11864](https://github.com/NVIDIA/NemoClaw/pull/11864).
  Refer to [Use NVIDIA Endpoints](/user-guide/openclaw/inference/hosted-inference/use-nvidia-endpoints).

## Documentation coverage

- Latest included cumulative docs PR: [#12247 — docs: prepare v0.0.129 documentation](https://github.com/NVIDIA/NemoClaw/pull/12247), merged 2026-09-23T20:59:46Z.
- Final PR commit: `be428a994e861cdaff7ecbf53b0f9a7f6018dc01`; merge commit: `26922313bba96184e65c3663b351683ebae9504d`.
- Final automated refresh coverage: `11541cde94ceb5fb96670628ed6ae18e3be76257`, derived from the verified workflow commit `e790c021d910299c46eede062bd7b98048d23908`.
- Later commit: `26922313bba96184e65c3663b351683ebae9504d`, [docs PR #12247](https://github.com/NVIDIA/NemoClaw/pull/12247). No intervening product commits.
- Changed paths: all eight are allowed documentation paths.
  - `docs/changelog/2026-09-23.mdx`
  - `docs/configure-agents/configure-memory-search.mdx`
  - `docs/deployment/install-openclaw-plugins.mdx`
  - `docs/inference/set-up-llama-cpp.mdx`
  - `docs/manage-sandboxes/gateway-lifecycle-control.mdx`
  - `docs/manage-sandboxes/recover-rebuild-sandboxes.mdx`
  - `docs/manage-sandboxes/runtime-controls.mdx`
  - `docs/reference/commands.mdx`
- Review decision: APPROVED. Independent documentation review cleared the final PR commit. Local docs build and publication validation passed.
- Open managed release-docs PRs: none at the historical-plan inspection. Later docs PR #12311 merged as `149fdc2e9fd19099e9907e6142cd4a3aed449e26`; it is outside this release range. Its later coverage does not alter the selected candidate.
- Maintainer decision: Proceed with the candidate as shown.
- Approved scope exception: tag the docs merge and use full E2E evidence from its parent; only the reviewed documentation patch differs.

### Documentation PR checks

| Check | State | Evidence |
|---|---|---|
| commit-lint | CANCELLED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918385089/job/107375731624) |
| request | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918379923/job/107375716536) |
| changes | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414956/job/107375836465) |
| assign-linked-issue-author | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918415122/job/107375835025) |
| label-release-target | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35919612126/job/107379923612) |
| changes | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384836/job/107375732327) |
| Analyze (go) | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918379858/job/107375720956) |
| Docs / Validate and publish PR preview | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384610/job/107375730049) |
| Docs / Verify CLI and installer reference parity | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384790/job/107375730793) |
| codebase-growth-guardrails | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918382156/job/107375723159) |
| codebase-growth-guardrails | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918415227/job/107375835909) |
| commit-lint | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918415038/job/107375928101) |
| require-maintainer-edits | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918381879/job/107375723036) |
| require-maintainer-edits | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918415131/job/107375835229) |
| dco-check | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384713/job/107375730829) |
| dco-check | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414968/job/107375833699) |
| CodeQL (javascript-typescript) | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384880/job/107375731157) |
| check-hash | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384734/job/107375730851) |
| check-hash | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414901/job/107375833207) |
| Analyze (javascript-typescript) | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918379858/job/107375720499) |
| CodeQL (python) | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384880/job/107375730940) |
| Analyze (python) | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918379858/job/107375721091) |
| docs-only-checks | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414956/job/107375883966) |
| docs-only-checks | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384836/job/107375807348) |
| ShellCheck SARIF | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384880/job/107375730696) |
| hugging-face-models | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384836/job/107375809790) |
| hugging-face-models | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414956/job/107375891448) |
| openshell-sdk-package | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384836/job/107375808901) |
| openshell-sdk-package | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414956/job/107375837857) |
| reviewed-npm-audit | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384836/job/107375809872) |
| reviewed-npm-audit | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414956/job/107375837445) |
| real-openclaw-dist-harness | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384836/job/107375809927) |
| real-openclaw-dist-harness | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414956/job/107375837237) |
| static-checks | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384836/job/107375810261) |
| static-checks | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414956/job/107375837677) |
| compile-artifacts | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384836/job/107375810391) |
| compile-artifacts | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414956/job/107375838221) |
| installer-integration | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384836/job/107375809985) |
| installer-integration | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414956/job/107375837866) |
| plugin-tests | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384836/job/107375811047) |
| plugin-tests | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414956/job/107375838003) |
| build-typecheck | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384836/job/107375811887) |
| build-typecheck | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414956/job/107375838769) |
| cli-test-shards | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384836/job/107375811940) |
| cli-test-shards | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414956/job/107375839487) |
| cli-tests | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384836/job/107375812408) |
| cli-tests | SKIPPED | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414956/job/107375840092) |
| checks | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918384836/job/107376436559) |
| checks | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/actions/runs/35918414956/job/107375838570) |
| CodeQL | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/runs/107376207040) |
| ShellCheck | SUCCESS | [Check](https://github.com/NVIDIA/NemoClaw/runs/107376035344) |
| CodeRabbit | SUCCESS | CodeRabbit skipped the bot-authored PR. |

The cancelled title-check run was superseded by a successful title check. Skipped runtime jobs are documented in the rollup; documentation checks passed.

## Base and managed image evidence

- Base-image candidate: `26922313bba96184e65c3663b351683ebae9504d`
- [Candidate push workflow](https://github.com/NVIDIA/NemoClaw/actions/runs/35919612485), attempt 1.
- [base-image-publication](https://github.com/NVIDIA/NemoClaw/actions/runs/35919612485/job/107379926224): completed successfully at 2026-09-23T21:01:29Z.
- Verified job and owning workflow SHA match the candidate; workflow path is `.github/workflows/e2e.yaml`, event is `push`, branch is `main`. The aggregate validates applicable base and managed-image publication.

## General E2E decision

- Maintainer selected full E2E run `35913145234` at `11541cde94ceb5fb96670628ed6ae18e3be76257` as release evidence.
- The candidate differs only by approved documentation PR #12247.
- Maintainer E2E decision: Proceed with the displayed failures and record release exceptions, as explicitly authorized after review of the failure history, merge suspects, and severity. The suite is not passing; this decision does not claim the failures are fixed or harmless.

### Requested full suite

- [Workflow](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234), attempt 1; tested commit `11541cde94ceb5fb96670628ed6ae18e3be76257`.
- Created: 2026-09-23T20:01:11Z. Final status inspected: 2026-09-23T21:25:16.369753+00:00.
- Status: completed; conclusion: failure.
- Twelve failed test jobs, including the now-finished staging Launchable; Release qualification also failed.

| Non-successful job or qualification | Status / conclusion | Start / completion (UTC) |
|---|---|---|
| [OpenShell credential generation window (docker)](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358819299) | completed / failure | 2026-09-23T20:03:59Z / 2026-09-23T20:11:05Z |
| [ubuntu-repo-cloud-langchain-deepagents-code: Repository install onboarding and hosted inference succeed [langchain-deepagents-code; Ubuntu managed-runtime host; NVIDIA hosted inference] [docker]](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358819300) | completed / failure | 2026-09-23T20:04:00Z / 2026-09-23T20:25:26Z |
| [MCP bridge (openclaw, docker)](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358820125) | completed / failure | 2026-09-23T20:03:59Z / 2026-09-23T20:17:03Z |
| [MCP bridge (hermes, docker)](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358820180) | completed / failure | 2026-09-23T20:04:00Z / 2026-09-23T20:14:52Z |
| [Exact staging Brev Launchable identity](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358820257) | completed / skipped | 2026-09-23T20:03:56Z / 2026-09-23T20:03:55Z |
| [native-runtime-qualification-producer-plan](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358820741) | completed / skipped | 2026-09-23T20:03:56Z / 2026-09-23T20:03:55Z |
| [jetson-nvmap-gpu](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358821122) | completed / skipped | 2026-09-23T20:03:56Z / 2026-09-23T20:03:56Z |
| [external-gateway-health](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358821736) | completed / skipped | 2026-09-23T20:03:56Z / 2026-09-23T20:03:55Z |
| [openshell-dev-artifact](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358821805) | completed / skipped | 2026-09-23T20:03:56Z / 2026-09-23T20:03:56Z |
| [Exact staging Brev Launchable](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358823997) | completed / failure | 2026-09-23T20:03:59Z / 2026-09-23T21:18:18Z |
| [Inference: Model Router returns a provider-routed response (docker) / NVIDIA API key](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358825509) | completed / failure | 2026-09-23T20:04:01Z / 2026-09-23T20:12:21Z |
| [Search: OpenClaw returns a Brave result without exposing its key (docker) / Brave and NVIDIA inference API keys](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358825609) | completed / failure | 2026-09-23T20:04:01Z / 2026-09-23T20:07:11Z |
| [Networking: OpenClaw answers through balanced egress (docker) / Brave and NVIDIA inference API keys](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358825802) | completed / failure | 2026-09-23T20:04:01Z / 2026-09-23T20:07:18Z |
| [Build pinned native Podman toolchain / ${{ matrix.architecture }}](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358825952) | completed / skipped | 2026-09-23T20:03:57Z / 2026-09-23T20:03:56Z |
| [Install: bootstraps NemoClaw and completes hosted inference (docker) / NVIDIA inference API key](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358826105) | completed / failure | 2026-09-23T20:04:01Z / 2026-09-23T20:05:37Z |
| [TUI: keeps rapid OpenClaw turns correlated (docker) / NVIDIA inference API key](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358826495) | completed / failure | 2026-09-23T20:04:01Z / 2026-09-23T20:07:28Z |
| [MCP bridge dev (${{ matrix.agent }}, ${{ matrix.runtime_provider }})](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358827485) | completed / skipped | 2026-09-23T20:03:57Z / 2026-09-23T20:03:57Z |
| [Authorization: approves a write-scope upgrade without operator.admin (docker) / NVIDIA inference API key](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358827629) | completed / failure | 2026-09-23T20:04:01Z / 2026-09-23T20:07:32Z |
| [matrix.jobName](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358836567) | completed / skipped | 2026-09-23T20:03:58Z / 2026-09-23T20:03:58Z |
| [Aggregate native runtime qualification evidence](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107358850463) | completed / skipped | 2026-09-23T20:04:01Z / 2026-09-23T20:04:00Z |
| [Protected managed-image GPU and local inference](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107363876532) | completed / failure | 2026-09-23T20:18:06Z / 2026-09-23T20:24:02Z |
| [Release qualification](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107386608916) | completed / failure | 2026-09-23T21:18:24Z / 2026-09-23T21:18:32Z |
| [Relevant E2E](https://github.com/NVIDIA/NemoClaw/actions/runs/35913145234/job/107386610281) | completed / skipped | 2026-09-23T21:18:19Z / 2026-09-23T21:18:19Z |

### Previously displayed full suite

- [Workflow](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340), attempt 1; tested commit `ad6ba127fb9eb0e52436699f45933bb0850d906b`.
- Created: 2026-09-22T22:15:06Z; started: 2026-09-22T22:15:06Z; last updated: 2026-09-22T23:34:24Z.
- Inspected: 2026-09-23T21:05:45+00:00; age from creation: 22 hours 50 minutes.
- Status: completed; conclusion: failure.

| Non-successful job or qualification | Status / conclusion | Start / completion (UTC) |
|---|---|---|
| [MCP bridge (openclaw, docker)](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961343514) | completed / failure | 2026-09-22T22:18:39Z / 2026-09-22T22:30:50Z |
| [MCP bridge (hermes, docker)](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961343524) | completed / failure | 2026-09-22T22:18:39Z / 2026-09-22T22:28:32Z |
| [Exact staging Brev Launchable identity](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961344018) | completed / skipped | 2026-09-22T22:18:37Z / 2026-09-22T22:18:36Z |
| [jetson-nvmap-gpu](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961344319) | completed / skipped | 2026-09-22T22:18:37Z / 2026-09-22T22:18:36Z |
| [native-runtime-qualification-producer-plan](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961344520) | completed / skipped | 2026-09-22T22:18:37Z / 2026-09-22T22:18:36Z |
| [openshell-dev-artifact](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961345048) | completed / skipped | 2026-09-22T22:18:37Z / 2026-09-22T22:18:36Z |
| [Exact staging Brev Launchable](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961345316) | completed / failure | 2026-09-22T22:18:39Z / 2026-09-22T23:33:42Z |
| [external-gateway-health](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961345947) | completed / skipped | 2026-09-22T22:18:37Z / 2026-09-22T22:18:36Z |
| [Networking: OpenClaw answers through balanced egress (docker) / Brave and NVIDIA inference API keys](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961346763) | completed / failure | 2026-09-22T22:18:40Z / 2026-09-22T22:21:45Z |
| [Search: OpenClaw returns a Brave result without exposing its key (docker) / Brave and NVIDIA inference API keys](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961346769) | completed / failure | 2026-09-22T22:18:40Z / 2026-09-22T22:21:44Z |
| [TUI: keeps rapid OpenClaw turns correlated (docker) / NVIDIA inference API key](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961347038) | completed / failure | 2026-09-22T22:18:40Z / 2026-09-22T22:22:15Z |
| [Install: bootstraps NemoClaw and completes hosted inference (docker) / NVIDIA inference API key](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961347196) | completed / failure | 2026-09-22T22:18:40Z / 2026-09-22T22:20:16Z |
| [Authorization: approves a write-scope upgrade without operator.admin (docker) / NVIDIA inference API key](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961347240) | completed / failure | 2026-09-22T22:18:40Z / 2026-09-22T22:22:51Z |
| [MCP bridge dev (${{ matrix.agent }}, ${{ matrix.runtime_provider }})](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961349124) | completed / skipped | 2026-09-22T22:18:38Z / 2026-09-22T22:18:37Z |
| [Build pinned native Podman toolchain / ${{ matrix.architecture }}](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961349466) | completed / skipped | 2026-09-22T22:18:38Z / 2026-09-22T22:18:37Z |
| [matrix.jobName](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961354719) | completed / skipped | 2026-09-22T22:18:39Z / 2026-09-22T22:18:38Z |
| [Aggregate native runtime qualification evidence](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106961357980) | completed / skipped | 2026-09-22T22:18:39Z / 2026-09-22T22:18:39Z |
| [Protected managed-image GPU and local inference](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106965356608) | completed / failure | 2026-09-22T22:34:50Z / 2026-09-22T22:43:21Z |
| [Release qualification](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106981685692) | completed / failure | 2026-09-22T23:33:46Z / 2026-09-22T23:33:53Z |
| [Relevant E2E](https://github.com/NVIDIA/NemoClaw/actions/runs/35791339340/job/106981686852) | completed / skipped | 2026-09-22T23:33:43Z / 2026-09-22T23:33:42Z |

### Accepted E2E exceptions and rationale

The maintainer explicitly instructed: "ok, lets put in exceptions for these and proceed to tag" after reviewing the following risks. Proceeding is a maintainer risk acceptance, not a passing qualification.

- Model Router: repeated 90-second completion timeout after successful onboarding. This affects normal requests through the optional routed provider. Root cause remains unresolved; no recent merge is strongly implicated. The maintainer accepts this functional risk for this release.
- Credential generation window: the pre-expiry probe reports denied. Its unchanged implementation also maps transport errors to denied, so authentication rejection is not established. The same OpenShell revision previously passed. Later expiry/revocation checks did not run; this lifecycle and security-validation gap is accepted.
- Deep Agents: onboarding, lifecycle and hosted inference passed; the newly added pinned-v1 consumer validation failed. PR #12161 added the failing shared validation path, while focused PR E2E covered OpenClaw/Hermes. Consumer build/environment failure versus export incompatibility remains unresolved. The maintainer accepts the export/migration compatibility gap.
- Eight recurring failures match September 22 signatures: MCP OpenClaw and Hermes connection probes; Brave export provider; networking preset provenance; empty TUI retry capture; authorization observer; bootstrap npm installation; and protected GPU/image-build package availability. Recurrence is not proof of harmlessness. These known failures and absent GPU qualification are accepted for this cut.
- Exact staging Brev Launchable was pending during the risk review and is now failed. Its job reports "full E2E failed"; no root cause or passing staging qualification is claimed. The final failed state is displayed for the final release confirmation and included in the exception scope.
- Historical comparisons used September 21 full run 35633262443, September 22 full run 35791339340, and September 23 push runs 35800188553, 35812859736, 35816051589, 35826347629, 35857794345, and 35908572283. The cited focused #12161 run 35899167950 passed OpenClaw and Hermes; Deep Agents was not selected. These contextual results do not replace the requested full run.

## Subsequent main evidence and historical selection

- The maintainer explicitly chose the original SHA after comparing the newer run and learning that newer main included product changes beyond that tested commit.
- Main at planning: `149fdc2e9fd19099e9907e6142cd4a3aed449e26`. The candidate remains its ancestor. Previous tag and peeled commit are unchanged.
- Excluded later PRs: #12093, #12235, #12302, #12303, #10384, #12307, #12237, and #12311. The last is the later docs merge; #12237 changed managed OpenClaw authentication behavior after the newer E2E SHA.
- The selected original candidate retains docs PR #12247 and its documentation-only difference from original E2E SHA `11541cde94ceb5fb96670628ed6ae18e3be76257`.
- The later standard-suite run [35939516667](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667), attempt 1, tested `8ed889c14ccd7e828a9fbc20606cc7c587e6ca31`. Created 2026-09-24T00:41:02Z; last updated 2026-09-24T01:05:21Z. Completed with failure: 67 successful, 14 failed, and 13 skipped workflow jobs. It excluded Launchable and skipped Release qualification. It is contextual evidence, not a passing full-suite rerun or evidence for the selected historical candidate.
- Compared with jobs executed in both runs, the later run had five newly failing jobs, one recovery, and nine repeated failures. Credential generation recovered. New failures were managed-image startup on AMD64 and ARM64 (missing contracts.json), cloud onboarding (legacy-file deletion assertion conflicting with new retention behavior), missing-sandbox repair (post-create finalization), and interrupted-onboarding resume (gateway health after restore). GPU qualification was skipped; this is not a recovery.
- Candidate image evidence was rechecked successfully while preparing this historical plan: original candidate push run35919612485, base-image-publication job107379926224. No later main check substitutes for that evidence.

### Later standard-suite failures

| Failed job | Evidence |
|---|---|
| MCP bridge (hermes, docker) | [Job](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667/job/107444664172) |
| Protected managed-image startup (linux/amd64) | [Job](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667/job/107444664187) |
| MCP bridge (openclaw, docker) | [Job](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667/job/107444664188) |
| Protected managed-image startup (linux/arm64) | [Job](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667/job/107444664259) |
| Cloud onboard (docker) | [Job](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667/job/107444664265) |
| ubuntu-repo-cloud-langchain-deepagents-code: Repository install onboarding and hosted inference succeed [langchain-deepagents-code; Ubuntu managed-runtime host; NVIDIA hosted inference] [docker] | [Job](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667/job/107444664306) |
| Networking: OpenClaw answers through balanced egress (docker) / Brave and NVIDIA inference API keys | [Job](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667/job/107444665398) |
| Search: OpenClaw returns a Brave result without exposing its key (docker) / Brave and NVIDIA inference API keys | [Job](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667/job/107444665405) |
| Onboarding: repairs a missing sandbox and rejects conflicting resume input (docker) / no provider credential | [Job](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667/job/107444665525) |
| Onboarding: resumes interrupted setup from recorded progress (docker) / no provider credential | [Job](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667/job/107444665736) |
| Inference: Model Router returns a provider-routed response (docker) / NVIDIA API key | [Job](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667/job/107444666337) |
| Authorization: approves a write-scope upgrade without operator.admin (docker) / NVIDIA inference API key | [Job](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667/job/107444666476) |
| Install: bootstraps NemoClaw and completes hosted inference (docker) / NVIDIA inference API key | [Job](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667/job/107444666602) |
| TUI: keeps rapid OpenClaw turns correlated (docker) / NVIDIA inference API key | [Job](https://github.com/NVIDIA/NemoClaw/actions/runs/35939516667/job/107444666660) |

Exceptions: The maintainer deliberately selects historical candidate `26922313bba96184e65c3663b351683ebae9504d` to retain the original reviewed release range and exclude later product changes and newly observed failures. The later standard E2E run remains failed and does not qualify this candidate. The maintainer accepts the reviewed docs-only difference between E2E SHA `11541cde94ceb5fb96670628ed6ae18e3be76257` and tag candidate `26922313bba96184e65c3663b351683ebae9504d`, and elects to ship with the explicitly documented Model Router, credential-window, Deep Agents compatibility, recurring E2E, and staging Launchable failures. Normal Deep Agents onboarding and inference passed, the other risks and validation gaps remain disclosed, and required candidate image publication passed. No full-suite success, security clearance, or fix is asserted.