- Tripy DebuggingDebug and diagnose errors in nvtripy code. Use when: interpreting TripyException stack traces, enabling MLIR/TensorRT debug output, understanding error reporting with stack_info, using raise_error, configuring debug environment variables, tracing compilation failures.NVIDIA/TensorRT-Incubator125
- Tripy DocumentationWrite API documentation for nvtripy following project conventions. Use when: writing docstrings for ops or modules, adding code examples, using @export.public_api document_under paths, creating Sphinx RST cross-references, understanding the docs build pipeline.NVIDIA/TensorRT-Incubator125
- Tripy New ModuleAdd a new neural network module to nvtripy. Use when: creating an nn layer, implementing a Module subclass, adding a new layer like Linear/LayerNorm/Conv, defining parameters with DefaultParameter or OptionalParameter, using constant_fields decorator.NVIDIA/TensorRT-Incubator125
- Tripy New OperationAdd a new operation to nvtripy. Use when: implementing a new op, adding a frontend op, creating a trace op, registering an op in the API. Covers the full Frontend → Trace → MLIR pipeline including export decorators, constraint definitions, and init registration.NVIDIA/TensorRT-Incubator125
- Tripy TestingWrite tests for nvtripy following project conventions. Use when: adding tests for ops, modules, trace operations, or compilation, using pytest parametrize, testing error cases with helper.raises, testing dtype combinations, understanding test directory structure.NVIDIA/TensorRT-Incubator125
- Nvalchemi Data StorageHow to write, read, compose, and load atomic data using nvalchemi's composable Zarr-backed storage pipeline (Writer, Reader, Dataset, MultiDataset, DataLoader). Use when saving simulation outputs or trajectories to disk, converting structures (e.g. ASE / extxyz) into Zarr stores, assembling datasets for training or inference, or wiring a DataLoader to stream batches to the GPU.NVIDIA/nvalchemi-toolkit106
- Nvalchemi Data StructuresHow to use AtomicData and Batch, the core graph-based data structures for representing atomic systems and batching them for GPU computation. Use when building systems from positions, cells, and atomic numbers, converting from ASE Atoms, batching or unbatching structures, reading per-atom vs per-graph tensors, or debugging shape, dtype, or device errors in model inputs.NVIDIA/nvalchemi-toolkit106
- Nvalchemi Dynamics ApiHow to configure and run dynamics simulations, compose multi-stage pipelines (FusedStage, DistributedPipeline), use inflight batching, and manage data sinks. Use when writing any simulation script — molecular dynamics (NVE/NVT), structure relaxation or geometry optimization (e.g. FIRE), equation-of-state or adsorption scans — or orchestrating many structures through a batched GPU pipeline.NVIDIA/nvalchemi-toolkit106
- Nvalchemi Dynamics HooksHow to use and write dynamics hooks — callbacks that observe or modify batch state at specific points during each simulation step. Use when a simulation needs neighbor-list rebuilds, convergence checks or early stopping, temperature control, per-step logging or trajectory capture, or any custom per-step behavior attached to a dynamics run.NVIDIA/nvalchemi-toolkit106
- Nvalchemi Dynamics ImplementationHow to implement a dynamics integrator by subclassing BaseDynamics and overriding pre_update() and post_update() methods. Use when creating a custom integrator, optimizer, or sampler that the built-in stages do not provide; for configuring existing dynamics, see nvalchemi-dynamics-api.NVIDIA/nvalchemi-toolkit106
- Nvalchemi Fine TuningHow to fine-tune nvalchemi-compatible models with FineTuningStrategy, pretrained checkpoint initialization, module patches, trainable-parameter filters, conservative optimizer defaults, validation, restart checkpoints, and model-agnostic MACE, AIMNet2, custom BaseModelMixin, or PyTorch inputs. Use when adapting a pretrained MLIP (e.g. MACE-MP) to new reference data, freezing or patching submodules during training, or resuming an interrupted fine-tune from a checkpoint.NVIDIA/nvalchemi-toolkit106
- Nvalchemi Loss ApiHow to use built-in loss functions and implement custom losses using the BaseLossFunction template-method pattern — residual types, per-atom normalization, masking, and graph-balanced reductions. Use when choosing or weighting energy, force, or stress objectives for training or fine-tuning, masking atoms or graphs out of the loss, or writing a custom loss term.NVIDIA/nvalchemi-toolkit106
- Nvalchemi Model WrappingHow to wrap an arbitrary MLIP (Machine Learning Interatomic Potential) using the BaseModelMixin interface to standardize inputs, outputs, and embeddings. Use when integrating a model such as MACE or AIMNet2 (e.g. MACEWrapper, loading pretrained checkpoints) so dynamics, training, or fine-tuning stages can call it, or when exposing energies, forces, or embeddings from a custom PyTorch model.NVIDIA/nvalchemi-toolkit106
- Nvalchemi ReportingHow to add observability to nvalchemi dynamics and training workflows using ReportingOrchestrator, RichReporter, TensorBoardReporter, scalar extraction, custom reporter callbacks, and dynamics LoggingHook. Use when showing live progress, writing TensorBoard summaries, preserving dynamics CSV rows, adding rank-safe distributed reporting, previewing Rich dashboards, or deciding between logging and reporting for training or molecular dynamics runs.NVIDIA/nvalchemi-toolkit106
- Nvalchemi Training ApiHow to configure nvalchemi training workflows with TrainingStrategy, custom training functions, standalone or composed losses, loss-weight schedules, optimizer and scheduler configs, validation, hooks, restartable checkpoints, and model-agnostic inputs. Use when training a model from scratch or setting up optimizers, schedulers, validation, or checkpointing for a training run; for adapting a pretrained model, see nvalchemi-fine-tuning.NVIDIA/nvalchemi-toolkit106
- Nvalchemi Zarr PerfPerformance tuning for nvalchemi's Zarr-backed Reader, Dataset, and DataLoader pipeline. Use when configuring AtomicDataZarrReader, Dataset, DataLoader, ZarrWriteConfig, or nvalchemi-io-test for training/inference throughput, especially shuffled access, graph-like random access, fused prefetch, pinned memory, validation overhead, or Zarr chunk/shard choices.NVIDIA/nvalchemi-toolkit106
- Compileiq Author ObjectiveUse when writing the objective_function= passed to Search(). Covers the two legal signatures (compiler-only str vs mixed list), the baseline-knockout branch, per-eval cache busting, framework-specific --apply-controls injection (raw PTXAS, NVCC, Triton, Helion, cuTeDSL/FA4, FlashInfer), correctness-before-timing, INVALID_SCORE handling, and the Debug-pack O0/O3 ACF-injection canary that must pass before launching a search. Triggers on "objective function", "apply-controls", "INVALID_SCORE", "savNVIDIA/CompileIQ103
- Compileiq Booster PackUse BEFORE running a full CompileIQ search. Walks through downloading a Booster Pack from NVIDIA/CompileIQ GitHub Releases, applying ACF candidates one at a time to the user's compiler (raw PTXAS, NVCC, Triton, Helion, FlashInfer), and keeping only candidates that compile, pass correctness, and beat the no-ACF baseline. Includes the mandatory Debug-pack O0/O3 ACF-injection canary that proves the ACF is reaching PTXAS. Triggers on "booster pack", "ACF", "apply-controls", "speed up without searchiNVIDIA/CompileIQ103
- Compileiq BootstrapUse when starting a fresh CompileIQ project, hitting a socket timeout, or before running any other compileiq-* skill. Verifies CUDA 13.3+, ptxas, GPU access, that `from compileiq.ciq import Search` and friends resolve, and that `PtxasSearchSpace().retrieve()` returns a real path. Documents the env vars that control timeouts, caching, and search-space mirroring. Triggers on "set up compileiq", "compileiq doesn't work", "socket timeout", "where do search spaces come from", "air-gapped compileiq".NVIDIA/CompileIQ103
- Compileiq DebugUse when something is wrong: Search() hangs, all evaluations return INVALID_SCORE, scores aren't improving, every config returns the same number, ptxas errors fill the log, CV% is too high, or a winning ACF candidate needs NCU profiling to explain. Symptom-indexed table on top. Triggers on "compileiq hang", "socket timeout", "INVALID_SCORE", "not converging", "every score is the same", "TypeError fromhex", "ncu profile", "register spill", "ptxas error", "not in expected format", "high cv".NVIDIA/CompileIQ103
- Compileiq Run SearchUse when composing the Search(...) call and calling .start(). Covers the four worker classes (MultiProcessWorker / IsoMultiProcessWorker / RayWorker / AsyncWorker) and when to pick each, SearchConfiguration sizing rules, dump_results checkpointing, tracker_config choice (Disabled / Loguru / MLflow), num_workers/task_timeout semantics, and GPU clock locking for stable measurements. Triggers on "Search()", "tuner.start()", "pool_size", "num_workers", "task_timeout", "IsoMultiProcessWorker", "RayWoNVIDIA/CompileIQ103
- Compileiq Search SpaceUse when picking the search_space= argument for Search(). Covers the three provider classes (PtxasSearchSpace, NvccSearchSpace, LocalSearchSpaceBin), how to pin a version/variant/tag, the attention-focused 'att' variant for attention kernels (FlashAttention, GQA, MHA, MLA, FlashInfer Batch Decode), air-gapped mirroring via CIQ_SEARCH_SPACES_DIR, and custom user-defined search spaces built from compileiq.search_spaces.base primitives. Triggers on "search space", "PtxasSearchSpace", "NvccSearchSpaNVIDIA/CompileIQ103
- Compileiq Validate ResultUse AFTER a Search has completed and BEFORE claiming any speedup or shipping an ACF. Loads the dump_results CSV, extracts top-K candidates (single-objective) or the Pareto front (multi-objective), re-measures each against the no-ACF baseline with 100+ trials on fresh caches, runs Welch's t-test plus Cohen's d, rejects three classic false-positive patterns (lucky-min / higher-variance / multiple-comparisons-of-N), and saves the validated winner as best.acf. Triggers on "validate result", "extractNVIDIA/CompileIQ103
- Update Rapids VersionUpdates python code (e.g. internal api calls) so that tests pass after running in conda environment with updated rapids version.NVIDIA/cuml-spark90
- Capture LearningAppend a learning to agent-learnings.md when the agent or user identifies a mistake, non-obvious pattern, or useful convention worth remembering. Use when the user says "capture this", "remember this", or "add to learnings".NVIDIA/multi-storage-client75
- Cloud Storage SetupSet up MSC credentials and MCP server integration for cloud storage access. Use when the user wants to interact with cloud storage (S3, GCS, Azure, OCI), asks to set up credentials, connect to a bucket, list objects, move files, or sync data. Checks for existing config first — only runs interactive setup if credentials are missing.NVIDIA/multi-storage-client75
- Fix CveFix CVE vulnerabilities by upgrading affected dependencies across Python (uv), Go, and Rust layers. Use when the user mentions CVE numbers, asks to fix security vulnerabilities, or asks to patch dependencies for CVEs.NVIDIA/multi-storage-client75
- Generate SpecProduce a feature specification covering API changes, backward compatibility, cross-language impact, performance, and a file-level change plan. Use when designing a new feature, planning a significant change, or the user asks to spec something out.NVIDIA/multi-storage-client75
- Generate Test PlanProduce a test plan with specific test names, assertions, and coverage across Python, Rust, and Go. Use when planning tests for a feature, after generating a spec, or the user asks for a test plan.NVIDIA/multi-storage-client75
- Prepare MrStage files, commit with correct format, push, and generate an MR description. Manual invocation only — has side effects (git operations). Use when the user explicitly asks to prepare, create, or submit an MR.NVIDIA/multi-storage-client75
- Release AutomationPrepares and automates the release process for multi-storage-client (version bump, release notes, checklist). Use when the user wants to cut a release, bump version, prepare release, publish a new version, or automate release steps.NVIDIA/multi-storage-client75
- ResearchBefore designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.NVIDIA/multi-storage-client75
- Add Binding FeatureAdd or change a public NeMo Relay API surface across the core runtime and every affected bindingNVIDIA/NeMo-Relay71
- Add MiddlewareAdd a new guardrail or intercept type to the NeMo Relay middleware pipelineNVIDIA/NeMo-Relay71
- Contribute ApiContribute a new NeMo Relay public API surface safely, with binding parity and docs in mindNVIDIA/NeMo-Relay71
- Contribute DocsContribute documentation or example changes that stay aligned with NeMo Relay public behaviorNVIDIA/NeMo-Relay71
- Contribute IntegrationContribute a new or updated third-party framework integration for NeMo RelayNVIDIA/NeMo-Relay71
- Karpathy GuidelinesBehavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.NVIDIA/NeMo-Relay71
- Maintain CiMaintain and review NeMo Relay GitHub Actions workflows with explicit per-job permissions, pinned action SHAs, deterministic caching, reusable workflow permission boundaries, and local validationNVIDIA/NeMo-Relay71
- Maintain Dynamic PluginsMaintain NeMo Relay dynamic plugin loaders, manifests, Rust native SDKs, gRPC worker protocol, Python worker SDK, docs, tests, and release workflow coverageNVIDIA/NeMo-Relay71
- Maintain ObservabilityMaintain or extend NeMo Relay observability surfaces across ATIF, OpenTelemetry, and OpenInferenceNVIDIA/NeMo-Relay71
- Maintain OptimizerMaintain or extend the NeMo Relay adaptive surface across config, plugins, docs, and bindings; use this when users still say optimizerNVIDIA/NeMo-Relay71
- Maintain PackagingMaintain NeMo Relay package metadata, module paths, generated artifacts, and release-facing build surfacesNVIDIA/NeMo-Relay71
- Nemo Relay Build PluginBuild and package reusable NeMo Relay runtime behavior as a config-activated plugin with validation and rollback-safe registrationNVIDIA/NeMo-Relay71
- Nemo Relay Debug Runtime IntegrationDebug application-side NeMo Relay integration issues such as load failures, inactive scopes, missing events, or adaptive/plugin wiring problemsNVIDIA/NeMo-Relay71
- Nemo Relay Export Atif TrajectoriesExport NeMo Relay activity as ATIF trajectories for replay, analysis, or interchangeNVIDIA/NeMo-Relay71
- Nemo Relay Export OpeninferenceConfigure and use NeMo Relay OpenInference export for OTLP backends that understand OpenInference semanticsNVIDIA/NeMo-Relay71
- Nemo Relay Export OtelConfigure and use NeMo Relay OpenTelemetry export for OTLP-compatible tracing backendsNVIDIA/NeMo-Relay71
- Nemo Relay Instrument CallsWrap application tool calls and LLM/provider calls with NeMo Relay scopes and managed execution APIsNVIDIA/NeMo-Relay71
- Nemo Relay Migrate From FlowMigrate applications, examples, integrations, documentation, package manifests, and repository code from NeMo Flow naming and packages to NeMo Relay across Python, Rust, Node.js, Go, C FFI, CLI, config, and observability surfaces; use when a user asks to rename nemo_flow/nemo-flow/NeMo Flow APIs, automate a migration, update imports or dependencies, or validate a Flow-to-Relay conversionNVIDIA/NeMo-Relay71
- Nemo Relay Setup ObservabilityChoose and set up the right NeMo Relay observability path for an applicationNVIDIA/NeMo-Relay71
- Nemo Relay StartHelp application developers pick a NeMo Relay binding and get to a first working scope, tool call, and LLM callNVIDIA/NeMo-Relay71
- Nemo Relay Tune Adaptive ConfigConfigure the NeMo Relay adaptive plugin component through the shared plugin system; use this skill for state, telemetry, adaptive_hints, tool_parallelism, ACG, or policy settings with validation and measured rolloutNVIDIA/NeMo-Relay71
- Nemo Relay Tune Adaptive HintsConsume NeMo Relay adaptive hints, predictions, latency sensitivity, ACG diagnostics, or tool-parallelism guidance safely in application logic after the adaptive plugin is already configuredNVIDIA/NeMo-Relay71
- Nemo Relay Tune PerformancePlan a measured NeMo Relay adaptive tuning rollout after baseline scopes, tool calls, LLM calls, and observability are working; use this skill to improve latency, tool parallelism, prompt-cache behavior, or model-request behavior from runtime signalsNVIDIA/NeMo-Relay71
- Nemo Relay Typed Wrappers CodecsUse NeMo Relay typed wrappers and codecs without losing middleware behaviorNVIDIA/NeMo-Relay71
- Nemo Relay Use Context IsolationSet up and reason about NeMo Relay scope-stack isolation for concurrent application workNVIDIA/NeMo-Relay71
- Prepare Code FreezePrepare a NeMo Relay code freeze by creating the release branch, updating nightly alpha branch config, bumping main to the next version, and opening the required PRNVIDIA/NeMo-Relay71
- Prepare PrPrepare, open, create, publish, update, or edit a NeMo Relay pull request or PR body with the right tests, docs, contributor hygiene, and repository pull request templateNVIDIA/NeMo-Relay71
- Rename SurfacesPerform a coordinated repository, package, crate, module, or symbol rename across NeMo RelayNVIDIA/NeMo-Relay71