From f6ae703974ea373d23e38d68977653406329e5b2 Mon Sep 17 00:00:00 2001 From: latenighthackathon Date: Mon, 25 May 2026 03:29:13 +0000 Subject: [PATCH] docs(reference): document remote-deploy env vars (closes #3059) Add a "Remote Deployment" subsection under Environment Variables in `docs/reference/commands.mdx` covering the four env vars that seed defaults for `nemoclaw deploy` and `nemoclaw onboard --remote`: - `NEMOCLAW_BREV_PROVIDER` - `NEMOCLAW_GPU` - `NEMOCLAW_DEPLOY_NO_CONNECT` - `NEMOCLAW_DEPLOY_NO_START_SERVICES` Each entry covers the default, the effect, and notes that flag equivalents on `deploy` take precedence when non-interactive runs need to skip prompts. The provider/GPU pair cross-links to the existing `Deploy to Remote GPU` narrative for full how-to coverage. The mirror at `.agents/skills/nemoclaw-user-reference/references/commands.md` is regenerated by `scripts/docs-to-skills.py` so the skill stays in sync with the user-facing docs. This is a fresh recreation of the closed #3349. The original branch was opened against the legacy `.md` docs layout before the Fern MDX migration (#3837) and accumulated 10 commits of upstream churn on `commands.mdx`; rebuilt cleanly on top of current upstream/main as a single signed commit. Signed-off-by: latenighthackathon --- .../nemoclaw-user-reference/references/commands.md | 13 +++++++++++++ docs/reference/commands.mdx | 13 +++++++++++++ 2 files changed, 26 insertions(+) diff --git a/.agents/skills/nemoclaw-user-reference/references/commands.md b/.agents/skills/nemoclaw-user-reference/references/commands.md index 48f4068c2f..07252482e9 100644 --- a/.agents/skills/nemoclaw-user-reference/references/commands.md +++ b/.agents/skills/nemoclaw-user-reference/references/commands.md @@ -1462,6 +1462,19 @@ These flags change defaults for commands that manage existing sandboxes. | `NEMOCLAW_DISABLE_INFERENCE_ROUTE_REPAIR` | `1` to enable | Skips the automatic DNS-proxy repair for stale `inference.local` routes during `nemoclaw connect` and `nemoclaw connect --probe-only`. Use only as a troubleshooting escape hatch. | | `NEMOCLAW_SHIELDS_ACCEPT_LEGACY_BASELINE` | `1` to opt in | Allows advanced immutable-config verification to trust the current on-disk bytes for older or partial content baselines. Use only after you have rebuilt or manually inspected the sandbox state and accepted that the baseline is operator-approved. | +### Remote Deployment + +These variables seed defaults for `nemoclaw deploy` and `nemoclaw onboard --remote`, which provision a sandbox on a Brev instance. +Each has a flag equivalent on `deploy`; the env var lets non-interactive runs skip the prompt. +For narrative how-to coverage of `NEMOCLAW_BREV_PROVIDER` and `NEMOCLAW_GPU`, see Deploy to Remote GPU. + +| Variable | Default | Effect | +|----------|---------|--------| +| `NEMOCLAW_BREV_PROVIDER` | `gcp` | Cloud provider for Brev instance creation. | +| `NEMOCLAW_GPU` | `a2-highgpu-1g:nvidia-tesla-a100:1` | GPU specification (instance type and GPU model) for the Brev instance. | +| `NEMOCLAW_DEPLOY_NO_CONNECT` | unset | When set to `1`, skips the automatic `connect` step after the remote deploy completes. | +| `NEMOCLAW_DEPLOY_NO_START_SERVICES` | unset | When set to `1`, skips starting services automatically after the remote deploy. | + ## NemoHermes Alias `nemohermes` is a convenience alias that pre-selects the Hermes agent. diff --git a/docs/reference/commands.mdx b/docs/reference/commands.mdx index 427df597ea..0025c618b9 100644 --- a/docs/reference/commands.mdx +++ b/docs/reference/commands.mdx @@ -1883,6 +1883,19 @@ These flags change defaults for commands that manage existing sandboxes. | `NEMOCLAW_DISABLE_INFERENCE_ROUTE_REPAIR` | `1` to enable | Skips the automatic DNS-proxy repair for stale `inference.local` routes during `nemoclaw connect` and `nemoclaw connect --probe-only`. Use only as a troubleshooting escape hatch. | | `NEMOCLAW_SHIELDS_ACCEPT_LEGACY_BASELINE` | `1` to opt in | Allows advanced immutable-config verification to trust the current on-disk bytes for older or partial content baselines. Use only after you have rebuilt or manually inspected the sandbox state and accepted that the baseline is operator-approved. | +### Remote Deployment + +These variables seed defaults for `nemoclaw deploy` and `nemoclaw onboard --remote`, which provision a sandbox on a Brev instance. +Each has a flag equivalent on `deploy`; the env var lets non-interactive runs skip the prompt. +For narrative how-to coverage of `NEMOCLAW_BREV_PROVIDER` and `NEMOCLAW_GPU`, see [Deploy to Remote GPU](../deployment/deploy-to-remote-gpu.md). + +| Variable | Default | Effect | +|----------|---------|--------| +| `NEMOCLAW_BREV_PROVIDER` | `gcp` | Cloud provider for Brev instance creation. | +| `NEMOCLAW_GPU` | `a2-highgpu-1g:nvidia-tesla-a100:1` | GPU specification (instance type and GPU model) for the Brev instance. | +| `NEMOCLAW_DEPLOY_NO_CONNECT` | unset | When set to `1`, skips the automatic `connect` step after the remote deploy completes. | +| `NEMOCLAW_DEPLOY_NO_START_SERVICES` | unset | When set to `1`, skips starting services automatically after the remote deploy. | + ### Legacy `nemoclaw setup` Deprecated. Use `nemoclaw onboard` instead.