← all publishers

NVIDIA

@nvidia source repo

Report this profile Claim this profile

• 962 skills
• 0 followers
• 2.2k repo stars
• 431 verified
• last week last updated

1. ▌
   Accessing Mlflow · nvidia

   Query and browse evaluation results stored in MLflow. Use when the user wants to look up runs by invocation ID, compare metrics across models, fetch artifacts (configs, logs, results), or set up the MLflow MCP server. ALWAYS triggers on mentions of MLflow, experiment results, run comparison, invocation IDs in the context of results, or MLflow MCP setup.
   2.2k repo stars
2. ▌
   Eagle3 New Model · nvidia

   Add a new model to the EAGLE3 offline pipeline. Generates an hf_offline_eagle3.yaml launcher config for a new model checkpoint, choosing the right hidden state dump backend (TRT-LLM / HF / vLLM) and GPU configuration. Use when user wants to run EAGLE3 on a model that does not yet have a YAML in tools/launcher/examples/ or asks how to configure the pipeline for a new checkpoint.
   2.2k repo stars
3. ▌
   Eagle3 Review Logs · nvidia

   Review EAGLE3 pipeline experiment logs from the launcher's experiments/ directory. Summarizes pass/fail status for all 4 tasks, diagnoses failures with root causes and fixes, and flags warnings. Use when the user asks to review job logs, check experiment results, or diagnose why a specific task failed.
   2.2k repo stars
4. ▌
   Quant Recipe Search · nvidia bundle

   Use when the user asks to find, search for, or optimize the best quantization recipe for a model, including direct requests like "find the best quantization recipe and generate a PTQ checkpoint." Guides the multi-candidate loop: choose compute-vs-memory success metrics, select ModelOpt recipe baselines, design AutoQuant/manual recipe deltas, interpret sensitivity, and decide next candidates. Do NOT use for a single known PTQ recipe run (use ptq), serving (use deployment), creating/running evals (use evaluation or launching-evals), monitoring jobs (use monitor), MLflow browsing (use accessing-mlflow), or comparing completed baseline-vs-candidate scores only (use compare-results).
   2.2k repo stars
5. ▌
   Release Cherry Pick · nvidia

   Cherry-pick merged PRs labeled for a release branch into that branch, then open a PR and apply the cherry-pick-done label. Use when asked to "cherry-pick PRs for release/X.Y.Z", "pick PRs to release branch", or "cherry-pick labeled PRs".
   2.2k repo stars
6. ▌
   Benchmark Model Kernels · nvidia bundle

   Inspect Hugging Face decoder layers on meta tensors and plan or run per-rank BF16, FP8, and NVFP4 GEMM or fused-MoE microbenchmarks with the bundled scripts and a local FlashInfer checkout. Use when choosing a model, GPU, TP, EP, or M/token-concurrency sweep; deriving common fused QKV and gate/up shapes without loading checkpoint weights; or using FlashInfer benchmark utilities. Do not use for end-to-end server throughput or request latency.
   2.2k repo stars
7. ▌
   Add Middleware · nvidia

   Add a new NeMo Relay guardrail or intercept type, registration surface, or pipeline stage. Do not use for changing an existing middleware implementation without a new middleware contract.
   2.2k repo stars
8. ▌
   Rename Surfaces · nvidia

   Rename a NeMo Relay repository, package, crate, module, public symbol, import path, or brand surface across multiple consumers. Do not use for a file-local private rename.
   2.2k repo stars
9. ▌
   Maintain Optimizer · nvidia

   Change NeMo Relay adaptive configuration, built-in adaptive components, reports, or binding helpers; also use when the task calls this surface optimizer. Do not use for generic plugin changes.
   2.2k repo stars
10. ▌
    Add Binding Feature · nvidia

    Add or change a public NeMo Relay runtime API across Rust and affected language bindings when no more specific maintainer skill owns the domain. Do not use for middleware additions, internal refactors, binding-local fixes, or docs-only changes.
    2.2k repo stars
11. ▌
    Nemo Relay Integrate Upstream · nvidia bundle

    Use this skill when assessing, extending, or implementing NeMo Relay support in an agent harness or agent framework, including coding agents and orchestration runtimes, when the host lacks Relay support or an existing integration needs deeper coverage. It identifies the host's execution boundaries and extension points, selects an appropriate attachment method for each boundary, and verifies the resulting coverage.
    2.2k repo stars
12. ▌
    Contribute Integration · nvidia

    Add or change a third-party framework integration that attaches NeMo Relay to framework tool, LLM, or lifecycle boundaries. Do not use for core runtime or binding-only changes.
    2.2k repo stars
13. ▌
    Maintain Observability · nvidia

    Change NeMo Relay event fields, subscribers, ATIF output, or typed OpenTelemetry and OpenInference projections. Do not use for instrumentation examples that leave observability contracts unchanged.
    2.2k repo stars
14. ▌
    Maintain Dynamic Plugins · nvidia

    Change NeMo Relay dynamic plugin loaders, manifests, native ABI, gRPC worker protocol, or Python worker SDK. Do not use for ordinary built-in plugin configuration.
    2.2k repo stars
15. ▌
    Small Fix · nvidia

    Make a small, reviewable NeMo Fabric bug fix without widening scope unnecessarily. Use for narrowly scoped bug fixes or behavior corrections in NeMo Fabric.
    2.2k repo stars
16. ▌
    Prepare Pr · nvidia

    Prepare, open, create, publish, update, or edit a NeMo Fabric pull request or PR body with the right tests, docs, scope, and review handoff details
    2.2k repo stars
17. ▌
    Maintain CI · nvidia

    Maintain and review NeMo Fabric GitHub Actions workflows with minimum permissions, pinned action SHAs, deterministic caching, lockfile-backed tools, and local validation
    2.2k repo stars
18. ▌
    Python Tests · nvidia

    Python tests for NeMo Fabric; use this when writing tests
    2.2k repo stars
19. ▌
    Contribute API · nvidia

    Contribute a new NVIDIA NeMo Fabric public API surface safely, with Rust, CLI, Python, TypeScript, schema, adapter, and documentation parity in mind
    2.2k repo stars
20. ▌
    Contribute Docs · nvidia

    Contribute documentation or example changes that stay aligned with NeMo Fabric public behavior
    2.2k repo stars
21. ▌
    Validate Change · nvidia

    Choose and run the right NeMo Fabric validation matrix for a change instead of using one fixed test list
    2.2k repo stars
22. ▌
    Review Doc Style · nvidia bundle

    Review NeMo Fabric documentation, examples, and docs-heavy changes for NVIDIA technical writing style, terminology, repository accuracy, and current public behavior. Use for documentation reviews and public-facing text changes.
    2.2k repo stars
23. ▌
    Contribute Adapter · nvidia

    Add or substantially change a first-party NVIDIA NeMo Fabric adapter, including repository packaging, discovery metadata, catalogs, CI wiring, and validation.
    2.2k repo stars
24. ▌
    Maintain Packaging · nvidia

    Maintain NVIDIA NeMo Fabric Rust, Python, and TypeScript dependencies, package metadata, module paths, native artifacts, lockfiles, license evidence, and release-facing build surfaces
    2.2k repo stars
25. ▌
    Draft Release Notes · nvidia bundle

    Compare NVIDIA NeMo Fabric release refs and draft the authoritative GitHub Release body plus any warranted documentation-site release-note update. Use when preparing a stable release, creating patch-release notes, updating docs/about-nemo-fabric/release-notes.mdx, or gathering verified release evidence.
    2.2k repo stars
26. ▌
    Prepare Code Freeze · nvidia

    Prepare a NeMo Fabric code freeze by creating a release branch, deciding whether frozen-line nightly alpha tags are required, bumping main to the next version, updating current-version documentation, and opening the required PR. Use when starting, preparing, or automating a NeMo Fabric code freeze.
    2.2k repo stars
27. ▌
    Update Project Version · nvidia

    Update the NVIDIA NeMo Fabric release version across Cargo, Python, and TypeScript package metadata, internal Python dependency pins, integration metadata, and lockfiles. Use when bumping, synchronizing, or auditing NeMo Fabric package versions for a release.
    2.2k repo stars
28. ▌
    Code Review · nvidia

    Review Python changes for correctness risks and missing test coverage.
    2.2k repo stars
29. ▌
    Swebench Debugging · nvidia

    Apply a focused reproduce-fix-test workflow to SWE-Bench repository tasks.
    2.2k repo stars
30. ▌
    Audit CI · nvidia bundle

    Run and analyze Elements cold CI performance profiles and propose evidence-backed build, test, lint, and dependency-graph improvements. Use whenever asked to profile or benchmark pnpm run ci, rerun ci:profile, create a CI performance audit, compare CI timings, find bottlenecks or the completion path, investigate a CI performance regression, or recommend measured CI/build optimizations.
    2.2k repo stars
31. ▌
    Audit Code · nvidia

    Comprehensive code review process for Elements monorepo changes. Provides structured feedback on type safety, testing, documentation, and adherence to project guidelines. Use this skill whenever the user asks you to review code, check staged changes, look at a diff, give feedback before committing, review a PR or merge request, or assess code quality. Trigger on phrases like "review these changes," "check this code," "give feedback," "look at the staged files," "review this PR," or "before committing."
    2.2k repo stars
32. ▌
    Agent Upkeep · nvidia bundle

    Perform one small, scoped maintenance improvement to the Elements monorepo and open a single reviewable pull request. Use this skill for scheduled or unattended upkeep runs that improve unit test coverage for one file, fix one behavioral bug in one module, or move one off ESLint rule toward enforcement to reduce type-safety debt. Trigger for requests to run upkeep, perform nightly maintenance, reduce technical debt, improve coverage for one file, improve type safety, enable one off lint rule, or when automation starts an agent with only a general codebase-maintenance goal.
    2.2k repo stars
33. ▌
    Agent Debug CI · nvidia

    Investigate failed scheduled or nightly GitHub Actions runs on the default branch, determine whether the failure is reproducible and repository-owned, implement the smallest evidence-backed fix, validate it, and open a pull request. Use when a failed nightly CI event starts an agent, or when the standard CI or Lighthouse jobs fail on the default branch and need an autonomous investigation and proposed PR fix.
    2.2k repo stars
34. ▌
    Authoring Tests · nvidia

    Write and run automated tests for Elements components including unit, accessibility, visual, SSR, and lighthouse tests. Use this skill whenever the user wants to write, create, update, or debug test files (.test.ts, .test.axe.ts, .test.visual.ts, .test.ssr.ts, .test.lighthouse.ts). Also trigger when the user asks about createFixture, removeFixture, elementIsStable, emulateClick, untilEvent, runAxe, visual baselines, theme testing, or test structure patterns like describe block naming.
    2.2k repo stars
35. ▌
    Troubleshooting · nvidia

    Diagnose and resolve common issues including test failures, build errors, performance regressions, and development environment problems. Use this skill whenever the user reports something broken, failing, timing out, hanging, or producing unexpected results. Trigger on test failures (elementIsStable timeouts, flaky tests, screenshot diffs), build errors (Cannot find module, wireit cache issues, stale output), CI/CD failures, Git LFS problems, port conflicts, lighthouse score regressions, SSR errors, or environment setup issues.
    2.2k repo stars
36. ▌
    Github Dependabot · nvidia

    Use the GitHub CLI to retrieve current open `Dependabot` alerts for a repository, inspect affected dependencies and patched versions, update manifests and lock files, and verify the local fixes. Use whenever a user asks to inspect, address, resolve, remediate, or fix `Dependabot` alerts or dependency security vulnerabilities reported by GitHub. Use `gh`, rather than a browser or GitHub connector, for alert discovery.
    2.2k repo stars
37. ▌
    Agent Dependencies · nvidia bundle

    Perform one conservative dependency or toolchain upgrade in the Elements monorepo and open a validated pull request. Use for nightly or unattended dependency maintenance, outdated package checks, first-party `actions/*` GitHub Actions updates, and pinned mise tool updates. Use the github-dependabot skill instead when the task starts from a GitHub security alert.
    2.2k repo stars
38. ▌
    Authoring Examples · nvidia

    Guidelines for writing example template files (*.examples.ts) including naming conventions, @summary JSDoc comments, and stateless HTML patterns. Use this skill whenever the user creates, modifies, or reviews *.examples.ts files, asks about example naming rules (PascalCase, 3-word max, no component prefix), needs to write or improve @summary JSDoc comments, encounters example-naming ESLint errors, or wants to add stateless or interactive examples to a component's example file.
    2.2k repo stars
39. ▌
    Authoring Patterns · nvidia

    Convert validated playground templates or HTML compositions into reusable pattern files (*.examples.ts) in the pattern library. Use this skill whenever the user wants to save, persist, store, or catalog a template, prototype, composition, or playground result as a reusable pattern. Trigger on phrases like "save as pattern," "create pattern," "add to pattern library," "persist this template," "convert to examples.ts," or when the user has validated HTML and wants it stored in projects/internals/patterns/src/.
    2.2k repo stars
40. ▌
    Authoring Projects · nvidia bundle

    Author a new standalone NVIDIA Elements workspace project containing a Lit component, including package files, the five test types, Wireit tasks, workspace and CI registration, metadata generation, documentation-site integration, dependency-lint configuration, and generated lock/notice files. Use when creating or scaffolding a new `projects/<name>` package based on the standalone component-library pattern. Do not use for adding a component to an existing project; use the authoring-components skill instead.
    2.2k repo stars
41. ▌
    Summarize Releases · nvidia bundle

    Create a monthly NVIDIA Elements “What’s New” docs page and a concise, copy-ready summary from local release tags, tagged changelogs, package versions, and commit history. Use for a scheduled release PR, monthly update, release roundup, changelog digest, Slack blurb, announcement, or plain-language explanation of recent Elements releases.
    2.2k repo stars
42. ▌
    Audit Accessibility · nvidia

    Unified accessibility auditing workflow across static analysis, runtime testing, ARIA patterns, keyboard navigation, and color contrast. Use this skill whenever the user mentions accessibility, a11y, WCAG, ARIA roles, axe tests, screen readers, focus management, keyboard navigation, color contrast, or wants to audit, verify, or fix accessibility on any component. Also use when writing or debugging .test.axe.ts files, checking tabindex management, or reviewing focus trapping behavior.
    2.2k repo stars
43. ▌
    Guidance API Design · nvidia

    Component API design patterns following Elements conventions for properties, attributes, CSS custom properties, slots, and events. Use this skill whenever the user is designing or deciding on a component API, choosing between properties vs attributes vs slots, naming CSS custom properties, designing events, avoiding impossible states, deciding whether to reflect to an attribute, using CSS Parts, implementing the DataElement interface, or applying the internal-host pattern. Also trigger when the user asks about shorthand vs granular CSS properties or event naming conventions.
    2.2k repo stars
44. ▌
    Authoring Benchmarks · nvidia

    Design, implement, run, debug, and interpret browser performance benchmarks for Elements components and utilities. Use whenever the user asks to benchmark or performance-test runtime code, create or update a .test.bench.ts file, compare benchmark results, investigate a browser performance regression, understand Vitest bench metrics such as throughput, mean, p99, RME, or samples, or add a test:bench task. Do not use this workflow for whole-CI profiling; use the audit-ci skill instead.
    2.2k repo stars
45. ▌
    Authoring Components · nvidia

    Guide for creating new Elements components with all required files, base classes, metadata patterns, and test boilerplate. Use this skill whenever the user wants to create, scaffold, or set up a new component from scratch, needs to understand the required 10-file structure, asks about base classes and mixins (LitElement vs ButtonFormControlMixin), define.ts vs index.ts patterns, static metadata, component registration, or sub-component parent relationships. Also trigger when the user mentions creating test boilerplate for all 5 test types.
    2.2k repo stars
46. ▌
    Authoring Typescript · nvidia

    Best practices for TypeScript code including type safety, discriminated unions, type guards, and exhaustive checking. Use this skill whenever the user asks about TypeScript patterns, type safety, type assertions (as any, as unknown), non-null assertions, discriminated unions, exhaustive switch/never patterns, type guards, boolean trap parameters, race conditions in async code, memory leaks from closures, or sequential vs parallel async operations. Also trigger when reviewing TypeScript code for unsafe patterns or refactoring types.
    2.2k repo stars
47. ▌
    Guidance Build · nvidia

    Wireit orchestration, build optimization, caching strategies, and build troubleshooting for the Elements monorepo. Use this skill whenever the user asks about Wireit configuration, build tasks, build dependencies, cache invalidation, build performance, adding new build tasks to package.json, cross-package dependency patterns, sideEffects declarations, CI pipeline configuration, or diagnosing slow or failing builds. Also trigger when the user mentions wireit, vite build issues, pnpm run ci, or build output problems.
    2.2k repo stars
48. ▌
    Authoring Documentation · nvidia

    Guidelines for writing documentation files including 11ty templates, Eleventy shortcodes, JSDoc annotations, and markdown content. Use this skill whenever the user works with documentation markdown files, Vale prose linting errors, Eleventy shortcodes (dodont, example), frontmatter, JSDoc annotations that must pass Vale, or the documentation site. Also trigger when the user mentions Vale errors, adding terms to the vocabulary, suppressing Vale rules, or writing and fixing prose in .md or .ts files.
    2.2k repo stars
49. ▌
    Agent Availability Report · nvidia bundle

    Generate a production availability report for NVIDIA Elements packages and documentation.
    2.2k repo stars
50. ▌
    Guidance Webgpu Performance · nvidia bundle

    Diagnose and improve browser WebGPU rendering performance, including frame pacing, draw and state-submission overhead, buffer uploads, shaders, and JavaScript work on the render critical path. Use for slow or stuttering GPU-backed interfaces, rendering regressions, CPU/GPU bottleneck analysis, or renderer performance changes; do not use for general page-load or build performance without a rendering symptom.
    2.2k repo stars
51. ▌
    Deep Research · nvidia bundle

    Queue deep, multi-step research and analysis tasks to the DeepAgents worker. Supports execution depth presets, request-specific rubrics, SubAgent delegation, and RubricMiddleware cross-validation.
    2.2k repo stars
52. ▌
    Nemoclaw Hermes Swarm · nvidia

    Use when deploying Hermes bots in NemoClaw sandboxes. Bring up, extend, verify, and debug a sandboxed bot swarm with ./swarm.
    2.2k repo stars
53. ▌
    Nemoclaw Community Maintainer Policies · nvidia bundle

    Provide read-only NemoClaw Community maintainer policy for Issue Type, labels, title conventions, triage, and example publication metadata.
    2.2k repo stars
54. ▌
    Nemoclaw Community Contributor Examples · nvidia bundle

    Guide NemoClaw Community contributors through classifying, naming, adding, moving, renaming, and reviewing repository examples. Use when work affects examples, recipes, partner contributions, field demos, developer tools, catalog collections, category indexes, upstream links, or example paths.
    2.2k repo stars
55. ▌
    Axe A11Y · nvidia bundle

    Automated web accessibility scanning (axe-core) driven by patchright + real Google Chrome — audits sites behind bot walls and behind auth
    2.2k repo stars
56. ▌
    Pr Review · nvidia

    Review one exact pull-request head with the read-only Hermes review advisor, an ordered evidence protocol, and a canonical finding ledger.
    2.2k repo stars
57. ▌
    Watchtower · nvidia bundle

    Run a scheduled web-surveillance sweep over a watchlist of topics using Tavily web_search, optional Tavily extraction, deterministic dedup/exclusion filtering, and a cited Markdown digest. Use when the user asks to run a sweep, run a watchtower sweep, monitor the watchlist, check the watchlist, or asks what changed since the last run. Trigger keywords - run a sweep, watchtower sweep, monitor watchlist, what changed.
    2.2k repo stars
58. ▌
    Module 1 · nvidia bundle

    This skill should be used when a learner is working through Module 1 ("Build an Agent") of the Build-an-Agent workshop and wants help understanding the concepts, notebooks, or code — e.g. "/module-1 what are agents?", "/module-1 explain the ReAct pattern", "help me with the docgen client exercise", "why use an agent instead of a single LLM call?", "my intro_to_agents notebook errors", "I'm stuck on Part 4 routing". It turns the agent into a Module 1 learning assistant (tutor) that explains concepts in the workshop's own framing, gives graduated hints WITHOUT ever completing exercises, interprets agent behavior, and troubleshoots the Module 1 notebooks/code/setup. Module 1 teaches the four agent components (model, tools, memory, routing), the agentic loop, the ReAct pattern, system prompts, and building a Report Generation Agent with NVIDIA Nemotron + Tavily + LangChain.
    2.2k repo stars
59. ▌
    Module 2 · nvidia bundle

    This skill should be used when a learner is working through Module 2 ("Agentic RAG") of the Build-an-Agent workshop and wants help understanding the concepts, code, or how to run it — e.g. "/module-2 what is agentic RAG?", "/module-2 explain chunking and embeddings", "help me with the reranker exercise", "how do I configure the MCP connection?", "my langgraph dev won't start", "AttributeError on 'ellipsis' object has no attribute 'split_documents'", "the agent won't use web search", "how do I migrate to a local NIM?". It turns the agent into a Module 2 learning assistant (tutor) that explains concepts in the workshop's own framing, gives graduated hints WITHOUT ever completing exercises or revealing the answer key, and troubleshoots the RAG agent, the LangGraph dev server, MCP, Skills, and the local-NIM migration. Module 2 builds an IT Help Desk agentic-RAG agent — chunk/embed (NeMo Retriever) into FAISS, rerank, expose retrieval as a tool, add web search via MCP, add dynamically loaded Skills, run it with `l
    2.2k repo stars
60. ▌
    Module 3 · nvidia bundle

    This skill should be used when a learner is working through Module 3 ("Agent Evaluation") of the Build-an-Agent workshop and wants help understanding the concepts, the code, or interpreting their results — e.g. "/module-3 what is faithfulness?", "/module-3 explain RAGAS metrics", "what's the difference between context precision and recall?", "my faithfulness score is 0.6, what does that mean?", "help me complete the FAITHFULNESS_PROMPT", "how do I read these evaluation results?", "is my LLM judge calibrated?", "RAGAS says not installed", "how do I generate an eval dataset?". It turns the agent into a Module 3 learning assistant (tutor) that explains evaluation concepts in the workshop's framing, gives graduated hints WITHOUT completing exercises or doing the learner's analysis for them, and troubleshoots the evaluation framework, datasets, judge model, and RAGAS. Module 3 builds an evaluation pipeline for the Module 1 (report) and Module 2 (RAG) agents using RAGAS metrics, LLM-as-a-judge, synthetic eval datas
    2.2k repo stars
61. ▌
    Module 4 · nvidia bundle

    This skill should be used when a learner is working through Module 4 ("Agent Customization") of the Build-an-Agent workshop and wants help understanding the concepts, the code, training, or GPU issues — e.g. "/module-4 should I train my agent or just prompt it?", "/module-4 what is GRPO?", "explain SFT vs GRPO", "how does the reward function work?", "what is reward hacking?", "help me with the GRPOConfig exercise", "my training crashes with OOM", "rewards aren't improving", "the reward server isn't responding", "what is NeMo Data Designer?", "how do I run the customized agent?". It turns the agent into a Module 4 learning assistant (tutor) that explains customization/RL concepts in the workshop's framing, gives graduated hints WITHOUT completing exercises or kicking off training runs, and troubleshoots SDG, the NeMo Gym reward server, GRPO/unsloth training, GPU/OOM, and the customized agent. Module 4 customizes a bash agent into a LangGraph CLI expert via synthetic data generation (NeMo Data Designer) + GRPO
    2.2k repo stars
62. ▌
    Module 5 · nvidia bundle

    This skill should be used when a learner is working through Module 5 ("Deep Agents") of the Build-an-Agent workshop and wants help understanding the concepts, the code, or sandboxing — e.g. "/module-5 what are deep agents?", "/module-5 explain the four pillars", "shallow vs deep agents?", "how does hierarchical delegation work?", "help me with the _build_backend exercise", "what's the difference between FilesystemBackend and LocalShellBackend?", "how does the Docker sandbox work?", "why isn't prompt-only security enough?", "my deep agent dry run fails", "the Deep Agents Client won't connect". It turns the agent into a Module 5 learning assistant (tutor) that explains deep-agent and sandboxing concepts in the workshop's framing, gives graduated hints WITHOUT completing exercises, models good security practice, and troubleshoots the demo backend, Docker sandbox, and the deepagents library. Module 5 builds a production deep agent (planning, delegation, memory, skills) with the deepagents library and Docker sandb
    2.2k repo stars
63. ▌
    Module 6 · nvidia bundle

    This skill should be used when a learner is working through Module 6 ("Agent Safety") of the Build-an-Agent workshop and wants help understanding the concepts, the code, or the NemoClaw stack — e.g. "/module-6 why isn't HITL or a container enough?", "/module-6 what does the Privacy Router actually do?", "explain Landlock / seccomp / the four layers", "what's the operator vs the agent?", "help me with the classify_sensitivity exercise", "how does the red-team runner score things?", "my nemoclaw sandbox won't connect", "the Live NemoClaw agent isn't the default". It turns the agent into a Module 6 learning assistant (tutor) that explains agent-safety concepts in the workshop's framing, gives graduated hints WITHOUT completing exercises, gets the Privacy Router's real behavior right, and troubleshoots the NemoClaw/OpenShell control plane and the safety-eval code. Module 6 hardens an autonomous OpenClaw agent with NVIDIA NemoClaw — kernel-level enforcement via OpenShell (network egress, Landlock filesystem, secco
    2.2k repo stars
64. ▌
    Module 7 · nvidia bundle

    This skill should be used when a learner is working through Module 7 ("Agent Harnesses & Skills") of the Build-an-Agent workshop and wants help understanding the concepts, the harness landscape, the lab code, or portable/GPU skills — e.g. "/module-7 what is a harness?", "/module-7 harness vs LLM?", "what's the context tax?", "explain lazy skill loading", "which harness should I use — pi vs OpenClaw vs Hermes vs Claude Code?", "help me with build_bare_agent", "my measure_context_tax numbers look off", "how do I write a portable SKILL.md?", "what are NVIDIA Verified Skills?", "how does the cuDF skill use my GPU?", "Hermes won't connect to Nemotron", "my self_evolve_skill exercise fails". It turns the agent into a Module 7 learning assistant (tutor) that explains harness/skills concepts in the workshop's framing, gives graduated hints WITHOUT completing exercises or revealing the answer keys, and troubleshoots the harness lab, Hermes, the verified-skill install, and the GPU exercise. Module 7 takes apart the har
    2.2k repo stars
65. ▌
    Workshop · nvidia bundle

    This skill should be used when a learner wants to navigate or understand the Build-an-Agent workshop as a whole — e.g. "/workshop where do I start?", "what order should I do the modules in?", "what does Module 3 need before I start?", "which module covers RAG / evaluation / training / safety?", "how do the modules connect?", "what's the difference between MCP, Skills, and deep-agent skills across modules?", "what have I finished and what's next?", "am I ready for the next module?". It is the workshop's overview + router that maps the seven-module arc and prerequisites, points the learner to the right /module-N skill, explains cross-module connections, and hosts the shared tutoring policy, glossary, and progress checks that all module skills draw on. For installing/launching the environment it points to the setup-workshop-nemoclaw skill (in-sandbox) and its setup-workshop-nemoclaw-operator half (host side).
    2.2k repo stars
66. ▌
    Release Packet · nvidia bundle

    Prepare a human-approval release packet for a screened payment. Does not release funds — release requires a human checker.
    2.2k repo stars
67. ▌
    Nv Tech Assistant · nvidia bundle

    Answer NVIDIA technical questions by searching authorized sources and citing retrieved evidence rather than relying on memory. Use for NVIDIA SDK or library usage and sample code, technology recommendations (including models and NGC/GitHub assets), adoption stories, troubleshooting, and product integration planning. Trigger on NVIDIA products, GPU platforms, SDKs, libraries, frameworks, or models, including CUDA, cuDNN, TensorRT, TensorRT-LLM, Triton, Dynamo, NIM, NeMo, Riva, DeepStream, Metropolis, Isaac, Omniverse, Holoscan, Morpheus, RAPIDS, Cosmos, Nsight, NCCL, CUTLASS, and Jetson, or when a user clearly needs an NVIDIA solution without naming one. Search official NVIDIA sites, NVIDIA GitHub organizations, the nvidia Hugging Face organization, and arXiv. Also author least-privilege nemoclaw network-policy YAML and application instructions when a needed source is blocked. Prefer this skill over answering NVIDIA specifics directly.
    2.2k repo stars
68. ▌
    Payment Screening · nvidia bundle

    Screen an outbound payment against limits, sanctions/watchlists, and duplicate checks before a human releases it.
    2.2k repo stars
69. ▌
    Rail Boundary Test · nvidia

    Verify that OpenShell denies FinGuard access to the synthetic payment rail. This is a control test, never a payment release.
    2.2k repo stars
70. ▌
    Payment Ops Playbook · nvidia

    The standard FinGuard payment-operations workflow — screen, explain, prepare, hand off to a human. Never release.
    2.2k repo stars
71. ▌
    Vss Video · nvidia

    Use when a question is about what happens in a video file or clip. Puts the clip in front of NVIDIA RT-VLM and reports timestamped, grounded findings.
    2.2k repo stars
72. ▌
    Retail API · nvidia bundle

    Araz Retail API — query stores, inventory, sales, customers, promotions, and orders. CLI: node ~/.openclaw/skills/retail-api/scripts/retail-api.js <command> --telegram-id TID. Authenticate with --telegram-id (the user's Telegram ID from message metadata from.id). Do NOT use --email or --password. Read commands: me, products [--category X] [--brand X] [--season X], inventory [--store N] [--product N] [--low-stock], customers [--id N] [--customer-email X], promotions [--all], sales [--store N], query "SQL". Write commands: transfer --product ID_OR_NAME --to-store N --quantity N [--size S] [--color C] [--from-store N], reorder --product ID_OR_NAME --quantity N [--store N] [--size S] [--color C]. QUERY SYNTAX: query "SELECT ... LIMIT 50" --telegram-id TID. The SQL is a positional argument — do NOT use -q flag (it does not exist). Table names are all lowercase: stores, products, inventory, orders, orderitems, customers, promotions, inventorytransfers, reorderrequests. Products PK is id, name column is product_name
    2.2k repo stars
73. ▌
    Pr Test Case Assistant · nvidia bundle

    Read public GitHub pull requests, produce a short triage brief, and draft feature test cases grounded in the pull request description and diff. Use when a Slack user provides a public GitHub repository or pull request.
    2.2k repo stars
74. ▌
    Setup Workshop Nemoclaw · nvidia bundle

    Set up the NVIDIA "Build an Agent" DevX workshop as a working JupyterLab environment from INSIDE a locked-down OpenShell/NemoClaw sandbox, and hand the user the token URL + access commands. This is the PRIMARY workshop-setup path in the NemoClaw community repo: use it when the user asks to set up / run / access the Build-an-Agent workshop and YOU are the agent running inside the sandbox. The workshop content is cloned from the upstream workshop repo to /sandbox/workshop-build-an-agent (setup.sh clones it if missing). NOT the workshop repo's own `setup-workshop` skill (a bare-metal GPU-host installer that needs sudo/Docker/CUDA and cannot run here), and NOT the host side — operators outside the sandbox use `setup-workshop-nemoclaw-operator`. This is the sandbox-native path: a scoped git clone, uv venv + pinned CPU deps, an LD_PRELOAD netlink shim to get Jupyter kernels past the seccomp AF_NETLINK block, a hand-built labextension bridge, launcher path rewrites, %pip-cell neutralization, and single-server discip
    2.2k repo stars
75. ▌
    Memory Repair · nvidia

    Check the memory against its schema and fix what can be fixed, so the store degrades visibly rather than silently.
    2.2k repo stars
76. ▌
    Memory Writing · nvidia

    Create and update the memory pages that the ranking job gates its top tier on, from evidence the selector collected.
    2.2k repo stars
77. ▌
    Inbound Judging · nvidia

    Decide which newly-arrived email and Slack messages create an obligation for the user, and craft the row that represents each one.
    2.2k repo stars
78. ▌
    Coach Nemoclaw Hermes · nvidia bundle

    Coach a NemoClaw Hermes sub-agent through Blender, OVRTX, OVPhysX, USD, rendering, simulation, and live Blender-control tasks. Use when Codex should delegate specialized execution to Hermes while managing sandbox boundaries, preventing overlapping runs, allowing tool iteration, intervening only on evidence, and independently validating artifacts.
    2.2k repo stars
79. ▌
    Nemoclaw Nvteam · nvidia bundle

    Route product, program, engineering, data and ML, quality, SRE, security, and developer-community work through the eight local role lenses packaged with the developer-community-chief-of-staff recipe in nemoclaw-community. Use for explicit NVTeam or persona activation, cross-functional readiness, developer relations, community enablement, technical-enablement work, or automatic specialist routing within this recipe. Do not use for a standalone question about core NemoClaw product capabilities unless the user explicitly requests NVTeam. This skill is Community-recipe behavior, not a built-in NemoClaw capability.
    2.2k repo stars
80. ▌
    Obligation Review · nvidia

    Re-judge open obligations against their current state in the world and re-rank them, on a schedule that does not depend on new mail arriving.
    2.2k repo stars
81. ▌
    Preference Update · nvidia

    Update a bounded preference policy from the user's repeated corrections, without ever overwriting the user's own choices.
    2.2k repo stars
82. ▌
    Source Etl Query · nvidia bundle

    Query the host-side source-etls REST mirror for GitHub discussions, historical GitHub mirror data, and NVIDIA forums research.
    2.2k repo stars
83. ▌
    Blender Host Sandbox Boundary · nvidia bundle

    Route Blender, USD, OVRTX, and OVPhysX work when Hermes terminal and file tools run in an OpenShell sandbox but Blender MCP runs inside a host Blender process. Use before any task that inspects the active scene, mentions host or sandbox paths, exports or validates files, or needs artifacts to cross the boundary.
    2.2k repo stars
84. ▌
    Ovphysx Host Runtime Boundary · nvidia

    Run configured native OVPhysX demos and report host GIF/status receipts when Hermes is sandboxed but the OVRTX/OVPhysX runtime and visible Blender process are on the host.
    2.2k repo stars
85. ▌
    Nemoclaw Autoheal · nvidia

    Guide users through Hermes availability checks and the optional host-side auto-heal setup without crossing the sandbox-to-host boundary.
    2.2k repo stars
86. ▌
    Setup Workshop Nemoclaw Operator · nvidia bundle

    Operator/host side of running the NVIDIA "Build an Agent" DevX workshop inside an OpenShell/NemoClaw sandbox. In the NemoClaw community repo this is the PRIMARY workshop-setup entry point: use it for "set up / run the Build-an-Agent workshop" requests when YOU are on the sandbox HOST (outside the sandbox — the machine running the OpenShell gateway + docker, e.g. via Claude Code) against this example's own vendored NemoClaw deployment (brought up by its scripts/bring-up.sh). It stages/applies the egress policy (PyPI, NIM /v1/ranking, scoped GitHub clone), stages this example's skills into the agent's skill library, optionally stages the NVIDIA API key, kicks the in-sandbox agent (which runs the `setup-workshop-nemoclaw` skill: clone + venv + JupyterLab), then opens the inbound path (openshell forward service + SSH/Teleport port-forward) so the user can open the JupyterLab token URL. Also covers sandbox lifecycle pitfalls (container restarts are only token-window-safe, recreate wipes state) and egress-denial de
    2.2k repo stars
87. ▌
    Memory Consolidation · nvidia

    Keep the memory inside its size limits by compacting rather than truncating, so it stays readable as it ages.
    2.2k repo stars
88. ▌
    Blender Python API Verification · nvidia bundle

    Verify Blender Python properties, enums, operators, animation APIs, and add-on interfaces against the running Blender build before writing or executing uncertain bpy code. Use for any Blender scene edit, render setup, animation, import/export, or add-on task where an API detail could vary by version.
    2.2k repo stars
89. ▌
    Github Readonly Live · nvidia bundle

    Read the configured live GitHub repository through authenticated, policy-scoped GitHub REST GET requests.
    2.2k repo stars
90. ▌
    Outlook Email Search · nvidia bundle

    Search the Outlook mailbox via Microsoft Graph to find and read emails that help answer user questions.
    2.2k repo stars
91. ▌
    Slack Channel Finder · nvidia bundle

    Discover Slack channels by topic, team, or domain and determine the channel ID.
    2.2k repo stars
92. ▌
    Coordinate Nemoclaw Blender · nvidia bundle

    Legacy compatibility entry point for prompts that explicitly request coordinate-nemoclaw-blender. Use coach-nemoclaw-hermes for new Codex-coached Blender, OVRTX, OVPhysX, USD, rendering, or simulation tasks.
    2.2k repo stars
93. ▌
    Slack Channel Summarizer · nvidia bundle

    Read and summarize Slack channel history from inside the NemoClaw sandbox.
    2.2k repo stars
94. ▌
    Cross Source Gap Analysis · nvidia

    Compare findings across Slack, GitHub, NVIDIA forums, and Outlook to identify alignment gaps, missing coverage, and follow-ups.
    2.2k repo stars
95. ▌
    Blackwall Payment Gate · nvidia

    Screen x402 payments with an advisory Blackwall verdict, then submit a payment intent to the host-side release gate. You prepare payments; only the gate can settle them.
    2.2k repo stars
96. ▌
    Public Web Search · nvidia

    Search the current public web through the recipe's optional, policy-scoped Tavily integration and return cited titles, URLs, snippets, and search metadata. Use for current public information that is not available from the configured Slack, Outlook, GitHub, or source-ETL skills. Do not use for opening a result page, extracting a URL, browser automation, or arbitrary HTTP requests.
    2.2k repo stars
97. ▌
    Gitlab Readonly Live · nvidia bundle

    Read one of the configured GitLab projects through authenticated, policy-scoped REST GET requests.
    2.2k repo stars
98. ▌
    Cutedsl Kernel Integration · nvidia bundle

    Use when integrating a CuTeDSL/CUTE DSL kernel into cuDNN Frontend as a frontend-only Python API, including APIBase wrappers, lazy cudnn exports, optional cutedsl dependencies, FE OSS documentation, and pytest coverage.
    2.2k repo stars
99. ▌
    Nvidia Kaggle Skill · nvidia bundle

    Use for Kaggle competition overview fetches, writeups, discussion/kernel research, submissions, and dataset uploads. Not for unrelated ML code.
    2.2k repo stars
100. ▌
     Simready Foundation Add Feature · nvidia bundle

     Use for adding SimReady feature docs, manifests, requirement mappings, validation strategy, and index entries.
     2.2k repo stars