maintain-ci

por nvidia

Manter e revisar os fluxos de trabalho do GitHub Actions do NeMo Fabric com permissões mínimas, SHAs de ações fixadas, cache determinístico, ferramentas baseadas em lockfile e...

npx skills add https://github.com/nvidia/nemo-fabric --skill maintain-ci

Maintain GitHub Actions CI

Companion Guidance

Use karpathy-guidelines alongside this skill for implementation or review work. Keep changes scoped, surface assumptions, and define focused validation before editing.

Use this skill when a change touches .github/workflows/*.yml or .github/workflows/*.yaml, or when reviewing CI behavior for security, reliability, or reproducibility.

Standards

  • Put permissions: on each job that needs token access.
  • Avoid workflow-level permissions unless the repository intentionally centralizes them and the inheritance tradeoff is documented.
  • Keep third-party actions pinned to full commit SHAs and preserve the readable version comment after the SHA.
  • Prefer action-native or ecosystem-native caching over generic actions/cache.
  • Use lockfiles or dependency manifests to drive cache invalidation.
  • Keep deploy and publish permissions isolated to the jobs that need them.
  • Publish the TypeScript contract from the dedicated publish_typescript.yml workflow through the protected npmjs environment. Grant id-token: write for npm trusted publishing, and do not provide an npm write token that could mask an OIDC configuration failure.
  • Read both caller and callee when a workflow uses workflow_call.
  • Keep documentation publish and preview credentials isolated to the Fern docs workflow.
  • Keep local commands aligned with the corresponding justfile recipes when they provide equivalent behavior.
  • Keep tag filters, prerelease normalization, and publication behavior aligned with RELEASING.md.

Permission Model

  • contents: read is the default minimum for checkout-based build, test, docs, and packaging jobs.
  • pull-requests: read is required for PR metadata lookup jobs.
  • pages: write should be limited to Pages deployment jobs and any caller that invokes them through a reusable workflow.
  • id-token: write should be limited to jobs that exchange a GitHub OIDC token with a protected deployment target, including Pages deployment and the protected npm publication job.
  • For reusable workflows, the caller must grant every permission the called jobs require. The callee cannot elevate beyond what the caller provides.

Caching

  • Prefer astral-sh/setup-uv cache support with cache-dependency-glob anchored to uv.lock.
  • Prefer Swatinem/rust-cache with explicit shared-key and workspaces instead of ad hoc target-directory caching.
  • Avoid caching generated outputs that can hide stale behavior unless the repo already relies on them deliberately.

Review Checklist

  • Each job has the minimum permissions it needs
  • Reusable workflow callers grant only the scopes their callees require
  • Every external action is pinned to a full SHA
  • Cache settings are tied to lockfiles, manifests, or explicit tool versions
  • Secrets are only passed to the jobs that consume them
  • Python, Rust, and documentation jobs remain aligned with their lockfiles and justfile recipes
  • Concurrency, branch filters, and documentation publish guards still reflect repository intent

Validation

Start with the narrowest useful checks:

just --fmt --check

Use ripgrep to inspect the workflow graph before editing:

rg -n "uses:|permissions:|secrets:|concurrency:|cache|just " .github/workflows

If local lint passes but the question is whether GitHub will authorize the run, inspect GitHub's permission model and the upstream action or reusable workflow source instead of assuming local success proves remote success.

Canonical References

  • .github/workflows/ci_python.yml
  • .github/workflows/ci_rust.yml
  • .github/workflows/fern-docs.yml
  • .github/workflows/nightly-alpha-tag.yml
  • .github/workflows/publish_rust.yml
  • .github/workflows/publish_typescript.yml
  • scripts/ci/publish_typescript_package.py
  • .gitlab-ci.yml
  • RELEASING.md
  • Cargo.lock
  • uv.lock
  • docs/package-lock.json
  • justfile
  • maintain-packaging
  • validate-change

Mais skills de nvidia

compileiq-debug
nvidia
Use quando algo está errado: Search() trava, todas as avaliações retornam INVALID_SCORE, as pontuações não estão melhorando, toda configuração retorna o mesmo número, erros de ptxas…
create-github-pr
nvidia
Crie pull requests do GitHub usando a CLI gh. Use quando o usuário quiser criar um novo PR, enviar código para revisão ou abrir um pull request. Palavras-chave de acionamento -…
nemoclaw-maintainer-cross-issue-sweep
nvidia
Escaneia outras issues abertas para encontrar aquelas que um determinado PR pode também corrigir ou quebrar acidentalmente. Gera oportunidades de correção adjacentes e riscos de contradição com arquivo:linha…
fhir-basics
nvidia
Ensina aos agentes como funcionam as APIs FHIR R4, quais recursos estão disponíveis, como consultá-los com parâmetros de busca e como analisar corretamente todos os formatos de resposta…
compileiq-validate-result
nvidia
Use APÓS a conclusão de uma Pesquisa e ANTES de reivindicar qualquer aceleração ou enviar um ACF. Carrega o CSV dump_results, extrai os K melhores candidatos (objetivo único)…
changelog-audit
nvidia
Auditar o CHANGELOG.md do Warp antes de um lançamento: recuperar entradas perdidas, ordenar por impacto ao usuário, refinar a linguagem das entradas, ajustar quebras de linha e (no modo de branch de lançamento) incrementar comparação…
maintain-dynamic-plugins
nvidia
Manter carregadores de plugins dinâmicos do NeMo Relay, manifestos, SDKs nativos em Rust, protocolo de worker gRPC, SDK de worker Python, documentação, testes e cobertura do fluxo de lançamento
dgx-diagnose
nvidia
Diagnostique problemas comuns do DGX Station GB300 — falhas de CUDA, direcionamento incorreto de GPU, bugs de contêiner vLLM/SGLang, problemas de estado MIG, erros de NVLink/Fabric Manager,…