Agent skills in plugin format
A skill is a Markdown file that tells a coding agent how to do one job. We parse every one we can find — the frontmatter, the outline, the languages in its code blocks — so you can look for the one that fits instead of reading through a repository to find out.
69,865 skills from 9,574 repositories, last read 11 September 2026.
Skills declare what they need inconsistently: 58.8% name the languages they run, 11.2% list the tools they ask for, and 4.6% give themselves a category. Filtering narrows to the ones that said so, never to the ones that are.
- nemo-mbridge-perf-megatron-fsdpNVIDIA/skills
- nemo-mbridge-perf-memory-tuningNVIDIA/skills
- nemo-mbridge-perf-moe-comm-overlapNVIDIA/skills
- nemo-mbridge-perf-moe-dispatcher-selectionNVIDIA/skills
- nemo-mbridge-perf-moe-hardware-configsNVIDIA/skills
- nemo-mbridge-perf-moe-long-contextNVIDIA/skills
- nemo-mbridge-perf-moe-optimization-workflowNVIDIA/skills
- nemo-mbridge-perf-moe-vlm-trainingNVIDIA/skills
- nemo-mbridge-perf-parallelism-strategiesNVIDIA/skills
- nemo-mbridge-perf-sequence-packingNVIDIA/skills
- nemo-mbridge-perf-tp-dp-comm-overlapNVIDIA/skills
- nemo-mbridge-recipe-recommenderNVIDIA/skills
- nemo-mbridge-resiliencyNVIDIA/skills
- nemo-relay-debug-runtime-integrationNVIDIA/skills
- nemo-relay-get-startedNVIDIA/skills
- nemo-relay-installNVIDIA/skills
- nemo-relay-instrument-callsNVIDIA/skills
- nemo-relay-instrument-context-isolationNVIDIA/skills
- nemo-relay-instrument-typed-wrappersNVIDIA/skills
- nemo-relay-migrate-from-flowNVIDIA/skills
- nemo-relay-plugin-adaptive-tuningNVIDIA/skills
- nemo-relay-plugin-buildNVIDIA/skills
- nemo-relay-plugin-observabilityNVIDIA/skills
- nemo-retrieverNVIDIA/skills
- nemo-rl-auto-researchNVIDIA/skills
- nemo-rl-brev-etiquetteNVIDIA/skills
- nemo-rl-docsNVIDIA/skills
- nemo-rl-session-memoryNVIDIA/skills
- nemoclaw-user-guideNVIDIA/skills
- nemotron-asr-finetuneNVIDIA/skills
- nemotron-customizeNVIDIA/skills
- nemotron-policy-generatorNVIDIA/skills
- nemotron-retrieval-recipesNVIDIA/skills
- nemotron-speechNVIDIA/skills
- nv-generate-ct-rflowNVIDIA/skills
- nv-generate-mr-brain-finetuneNVIDIA/skills
- nv-generate-mr-brainNVIDIA/skills
- nv-generate-mrNVIDIA/skills
- nv-generate-vae-finetuneNVIDIA/skills
- nv-reason-cxrNVIDIA/skills
- nv-segment-ct-finetuneNVIDIA/skills
- nv-segment-ctNVIDIA/skills
- nv-segment-ctmrNVIDIA/skills
- nvidia-skill-finderNVIDIA/skills
- omniverse-cad-to-simreadyNVIDIA/skills
- omniverse-realtime-viewerNVIDIA/skills
- omniverse-usd-performance-tuningNVIDIA/skills
- paidf-anomalygenNVIDIA/skills
- paidf-augmentationNVIDIA/skills
- paidf-auto-labelingNVIDIA/skills
- paidf-curation-and-retrievalNVIDIA/skills
- paidf-orchestration-setupNVIDIA/skills
- paidf-orchestration-write-dagNVIDIA/skills
- physical-ai-defect-image-generationNVIDIA/skills
- physical-ai-event-video-generationNVIDIA/skills
- physical-ai-image-attribute-augmentationNVIDIA/skills
- physical-ai-infrastructure-setup-and-resilient-scalingNVIDIA/skills
- physical-ai-neural-reconstructionNVIDIA/skills
- physical-ai-video-data-augmentationNVIDIA/skills
- physicsnemo-discoverNVIDIA/skills
- physicsnemo-shard-tensorNVIDIA/skills
- portfolio-optimizationNVIDIA/skills
- rag-blueprintNVIDIA/skills
- rag-evalNVIDIA/skills
- rag-perfNVIDIA/skills
- rtvi-cv-customize-modelNVIDIA/skills
- rtvi-cv-scaffold-vss-serviceNVIDIA/skills
- rtvi-vlm-customize-modelNVIDIA/skills
- skill-card-generatorNVIDIA/skills
- tao-analyze-changenet-rcaNVIDIA/skills
- tao-analyze-gaps-visual-changenetNVIDIA/skills
- tao-analyze-gaps-vlm-bcqNVIDIA/skills
- tao-convert-dataset-formatNVIDIA/skills
- tao-finetune-clipNVIDIA/skills
- tao-finetune-cosmos-embedNVIDIA/skills
- tao-finetune-cosmos-reasonNVIDIA/skills
- tao-finetune-huggingface-modelNVIDIA/skills
- tao-generate-image-groundingNVIDIA/skills
- tao-generate-referring-expressionsNVIDIA/skills
- tao-generate-video-reasoning-annotationsNVIDIA/skills
- tao-launch-workflowNVIDIA/skills
- tao-list-capabilitiesNVIDIA/skills
- tao-mine-aoi-imagesNVIDIA/skills
- tao-port-huggingface-modelNVIDIA/skills
- tao-route-visual-changenet-samplesNVIDIA/skills
- tao-run-automl-deft-pipelineNVIDIA/skills
- tao-run-automlNVIDIA/skills
- tao-run-deft-aoiNVIDIA/skills
- tao-run-inference-serviceNVIDIA/skills
- tao-run-on-brevNVIDIA/skills
- tao-run-on-dockerNVIDIA/skills
- tao-run-on-kubernetesNVIDIA/skills
- tao-run-on-local-dockerNVIDIA/skills
- tao-run-on-slurmNVIDIA/skills
- tao-run-platformNVIDIA/skills
- tao-setup-nvidia-gpu-hostNVIDIA/skills
- tao-train-action-recognitionNVIDIA/skills
- tao-train-bevfusionNVIDIA/skills
- tao-train-centerposeNVIDIA/skills
- tao-train-deformable-detrNVIDIA/skills
- tao-train-depth-anything-v2NVIDIA/skills
- tao-train-dinoNVIDIA/skills
- tao-train-fast-foundation-stereoNVIDIA/skills
- tao-train-foundation-stereoNVIDIA/skills
- tao-train-grounding-dinoNVIDIA/skills
- tao-train-image-classificationNVIDIA/skills
- tao-train-mask-auto-encoderNVIDIA/skills
- tao-train-mask-auto-labelNVIDIA/skills
- tao-train-mask-grounding-dinoNVIDIA/skills
- tao-train-mask2formerNVIDIA/skills
- tao-train-metric-learning-recognitionNVIDIA/skills
- tao-train-nvdinov2NVIDIA/skills
- tao-train-nvpanoptix3dNVIDIA/skills
- tao-train-ocdnetNVIDIA/skills
- tao-train-ocrnetNVIDIA/skills
- tao-train-oneformerNVIDIA/skills
- tao-train-optical-inspectionNVIDIA/skills
- tao-train-pointpillarsNVIDIA/skills
- tao-train-pose-classificationNVIDIA/skills
- tao-train-reidNVIDIA/skills
- tao-train-rtdetrNVIDIA/skills
- tao-train-segformerNVIDIA/skills
- tao-train-single-stepNVIDIA/skills
- tao-train-sparse4dNVIDIA/skills
- tao-train-visual-changenetNVIDIA/skills
- tao-validate-dataset-formatNVIDIA/skills
- tilegym-adding-cutile-kernelNVIDIA/skills
- tilegym-converting-cutile-to-juliaNVIDIA/skills
- tilegym-converting-cutile-to-tritonNVIDIA/skills
- tilegym-cutile-autotuningNVIDIA/skills
- tilegym-cutile-pythonNVIDIA/skills
- tilegym-improve-cutile-kernel-perfNVIDIA/skills
- tilegym-monkey-patch-kernels-to-transformersNVIDIA/skills
- vss-ask-videoNVIDIA/skills
- vss-deploy-dense-captioningNVIDIA/skills
- vss-deploy-detection-tracking-2dNVIDIA/skills
- vss-deploy-detection-tracking-3dNVIDIA/skills
- vss-deploy-profileNVIDIA/skills
- vss-deploy-video-embeddingNVIDIA/skills
- vss-generate-video-calibrationNVIDIA/skills
- vss-generate-video-reportNVIDIA/skills
- vss-manage-alertsNVIDIA/skills
- vss-manage-video-io-storageNVIDIA/skills
- vss-query-analyticsNVIDIA/skills
- vss-search-archiveNVIDIA/skills
- vss-setup-behavior-analyticsNVIDIA/skills
- vss-setup-video-analytics-apiNVIDIA/skills
- vss-summarize-videoNVIDIA/skills
- warp-compile-time-optimizerNVIDIA/skills
- warp-debug-gradientsNVIDIA/skills
- warp-evalNVIDIA/skills
- memory-bankNagendhra-web/memory-bank
- log-analysisNailShakurov/logzip
- sepia-hemingwayNanako0129/sepia
- sepia-recreateNanako0129/sepia
- sepia-refactorNanako0129/sepia
- sepia-reviewNanako0129/sepia
- sepia-writeNanako0129/sepia
- sepiaNanako0129/sepia
- tapestryNatsuFox/Tapestry
- mineruNebutra/MinerU-Skill
- mineruNebutra/MinerU-Skill
- ax-extract-workflowNecmttn/ax
- ax-repoNecmttn/ax
- dojoNecmttn/ax
- efficient-dispatchNecmttn/ax
- ax-narrateNecmttn/ax
- release-announcementNecmttn/ax
- retro-metaNecmttn/ax
- retroNecmttn/ax
- setupNecmttn/ax
- ship-checklistNecmttn/ax
- 18f-contentNeeeophytee/agent-stylebooks
- apple-interface-writingNeeeophytee/agent-stylebooks
- cdc-clear-communicationNeeeophytee/agent-stylebooks
- github-docsNeeeophytee/agent-stylebooks
- gitlab-docsNeeeophytee/agent-stylebooks
- google-developer-docsNeeeophytee/agent-stylebooks
- govukNeeeophytee/agent-stylebooks
- kubernetes-docsNeeeophytee/agent-stylebooks
- mailchimp-contentNeeeophytee/agent-stylebooks
- mdn-web-docsNeeeophytee/agent-stylebooks
- microsoft-writing-styleNeeeophytee/agent-stylebooks
- nasa-technical-writingNeeeophytee/agent-stylebooks
- nhs-health-contentNeeeophytee/agent-stylebooks
- red-hat-docsNeeeophytee/agent-stylebooks
- sec-plain-englishNeeeophytee/agent-stylebooks
- w3c-technical-reportsNeeeophytee/agent-stylebooks
- advisor-call-budgetNeeeophytee/ai-cost-cutter-skills
- cheap-swap-guardNeeeophytee/ai-cost-cutter-skills
- context-dietNeeeophytee/ai-cost-cutter-skills
- free-model-triageNeeeophytee/ai-cost-cutter-skills
- free-tier-batch-planNeeeophytee/ai-cost-cutter-skills
- model-bakeoffNeeeophytee/ai-cost-cutter-skills
- reasoning-effort-throttleNeeeophytee/ai-cost-cutter-skills
- route-cheap-escalate-hardNeeeophytee/ai-cost-cutter-skills
- tested-fallbackNeeeophytee/ai-cost-cutter-skills
- token-receipts-auditNeeeophytee/ai-cost-cutter-skills
- agent-interface-designNeeeophytee/finding-unknowns-skills
- blindspot-passNeeeophytee/finding-unknowns-skills
The whole corpus is one call away through the search API, or ask a question of it on the index itself.