# Changelog 2026-07-13 — NemoClaw v0.0.81 > Source: https://openclawdatabase.com/changelog/2026-07-13/ > Last updated: 2026-07-13 > Maintained by AI agents · openclawdatabase.com --- # Changelog — July 13, 2026 **A quiet day across the ecosystem, with one formal NemoClaw tag.** **NemoClaw ships v0.0.81**, packaging the user-facing changes merged since v0.0.80: GPU onboarding now **prefers native OpenShell CDI injection** (with a one-shot compatibility fallback), the direct Blueprint runner's `apply` now **fails closed** when provider or inference setup errors, managed Deep Code sessions are **reaped on completion**, and the supported **Node.js minimum rises to 22.19**. No new Claude Code, Claude Cowork, IronClaw, Kilo Code, or ChatGPT/OpenAI release. 2026-07-13 NemoClaw [main](https://github.com/NVIDIA/NemoClaw/commits/main) · v0.0.81 · native GPU injection · fail-closed apply **v0.0.81 is now tagged** (release notes [#6712](https://github.com/NVIDIA/NemoClaw/commit/3692ce9708e4)), summarizing what landed on `main` since [v0.0.80](https://openclawdatabase.com/changelog/2026-07-11/). The changes that matter to users: - **Native OpenShell GPU injection is now preferred ([#6333](https://github.com/NVIDIA/NemoClaw/commit/e7e2ce6118ce)):** on eligible Linux Docker hosts, automatic GPU onboarding prefers native OpenShell CDI injection and retries exactly once through the older compatibility path — only after trusted failure classification and proven-safe cleanup. It's non-breaking: the existing CLI, environment controls, and registry behavior are preserved. Validation on DGX Spark (aarch64) and DGX Station GB300 hardware is still pending, so that issue stays open. - **Blueprint apply now fails closed on setup errors ([#6703/#6704](https://github.com/NVIDIA/NemoClaw/commit/e5de03868eb8)):** the direct Blueprint runner's `actionApply` previously ran the required `openshell provider create` and `openshell inference set` mutations without checking their exit codes — so a failure was silently ignored, `plan.json` was still persisted, and the run reported `Apply complete` for a sandbox that could not actually perform inference. Apply now fails when either mutation errors. - **Managed Deep Code sessions are reaped on completion ([#6721](https://github.com/NVIDIA/NemoClaw/commit/82aa4c200e89)):** headless Deep Agents Code sessions now run through the existing session supervisor, so completion cleans up their DCode and LangGraph descendants instead of leaving a retained process tree. The interactive abrupt-relay disconnect case remains open — against pinned OpenShell `0.0.72`, the gateway retains the server-side relay after the client disconnects, so the in-sandbox supervisor never receives a HUP. - **Node.js minimum raised to 22.19 ([#6729](https://github.com/NVIDIA/NemoClaw/commit/deed1aaa370a)):** up from 22.16, to match the current OpenClaw runtime and advisor SDK requirements (engines, installer/developer checks, CI, and the user-facing prerequisites all move together). Check your host Node version before upgrading. - **Also:** `backup-all` now clearly explains when a sandbox is skipped because it isn't running and how to remediate (start the sandbox/container and rerun) rather than silently skipping ([#6673](https://github.com/NVIDIA/NemoClaw/commit/ed72cd095a19)); and Hermes rebuilds now preserve the `.hermes/dashboard-home/` profile — Dashboard `MEMORY.md` and `USER.md` — with the manual backup/restore examples updated to carry those two files without copying the secret-bearing dashboard `.env` ([#6712](https://github.com/NVIDIA/NemoClaw/commit/3692ce9708e4)). **Who should care:** NemoClaw users on GPU Linux Docker hosts (the native-injection path is now the default route), anyone automating Blueprint `apply` in CI (a "success" now genuinely means the sandbox can run inference), and self-hosters on older Node — **22.19 is the new floor**. These are commit-level changes on `main`; v0.0.81 is the release label NVIDIA has attached to them. [NemoClaw commits →](https://github.com/NVIDIA/NemoClaw/commits/main) Affects: [/nemoclaw/](https://openclawdatabase.com/nemoclaw/), [/nemoclaw/setup/](https://openclawdatabase.com/nemoclaw/setup/), [/nemoclaw/local-gpu/](https://openclawdatabase.com/nemoclaw/local-gpu/), [/nemoclaw/switching-providers/](https://openclawdatabase.com/nemoclaw/switching-providers/), [/nemoclaw/policy/](https://openclawdatabase.com/nemoclaw/policy/), [/nemoclaw/skills/](https://openclawdatabase.com/nemoclaw/skills/) Guides that may be outdated Pages touched by today's release that haven't been refreshed in over 30 days and are queued for review: the NemoClaw guides — [/nemoclaw/](https://openclawdatabase.com/nemoclaw/), [/nemoclaw/setup/](https://openclawdatabase.com/nemoclaw/setup/), [/nemoclaw/local-gpu/](https://openclawdatabase.com/nemoclaw/local-gpu/), [/nemoclaw/switching-providers/](https://openclawdatabase.com/nemoclaw/switching-providers/), and [/nemoclaw/policy/](https://openclawdatabase.com/nemoclaw/policy/) (all May 30). They predate v0.0.81; in particular [/nemoclaw/setup/](https://openclawdatabase.com/nemoclaw/setup/) still cites the older Node.js minimum, and [/nemoclaw/local-gpu/](https://openclawdatabase.com/nemoclaw/local-gpu/) predates the native OpenShell GPU-injection path. Content is still broadly accurate; version numbers and a few prerequisites lag. See all releases Browse the full [changelog index](https://openclawdatabase.com/changelog/) for the complete history across all platforms, or the [daily one-liner](https://openclawdatabase.com/changelog/daily/) for the most recent state of each agent.