Data analysis, machine learning, LLM apps, and prompt design (9.1K skills)
Build automated data checks with Great Expectations and dbt to detect missing values, duplicates, invalid ranges, and stale records. Define data contracts, shared agreements on data structure and quality, and summarize validation results across tables.
Turn analytical findings into reports and presentations that explain what the numbers mean and what to do next. Structure the narrative, choose clear visual comparisons, highlight key findings, and communicate uncertainty to audiences without technical expertise.
Learn how to organize the instructions, conversation history, documents, and tool results an AI agent receives. Explains context limits and loading information as needed, helping you review agent designs and investigate behavior caused by poorly managed context.
Add language model features to a PinMe TypeScript Worker through its OpenRouter proxy. The skill guides code for listing models, chat responses, streaming text as it arrives, web search, and handling API errors.
Improve a Browserbase Agent's system prompt by running a fixed task, examining messages and logs, and changing one instruction at a time. Compare quality, duration, and message counts, then repeat runs to check consistency.
Generate synthetic printed circuit board inspection images through a CDF workflow. Choose simulation, realistic clean images, defect generation, or the full pipeline, then validate settings, check GPU readiness, monitor execution, and trace defects to their source images.
Create and edit Proteina-Complexa target definitions for protein binders, ligand binders, and enzyme scaffolding. Organize chain and residue selections in YAML configuration files, check saved settings, and verify structure file paths when a prepared installation is available.
Compare Proteina-Complexa design settings across all combinations of selected parameter values. Prepare sweep configurations, estimate GPU time, coordinate runs, and produce ranked CSV results that show success rates and trade-offs between quality and runtime.
Generate videos of unusual events, such as falls or fires, from seed images. Run a workflow that adds object detection, tracking, captions, and other labels, then builds an anomaly dataset and retrieves the results.
Run the CDF Video Data Augmentation workflow to create Cosmos video variants and automatically label source and generated videos. Prepare and validate run settings, check available resources, monitor execution, and verify the separate training exports.
Build validated simulation jobs from a board USD, a 3D scene file, for clean board images, pin variations, component placement defects, or solder defects. Return the compiled job payload for a separate executor to render.
Evaluate existing protein structure files with Proteina-Complexa, including outputs from BindCraft, AlphaProteo, and RFdiffusion. Select a structure prediction backend, calculate interface and structural agreement metrics, and summarize individual designs, pass rates, and leading candidates in CSV outputs.
Create clothing variations from cropped person images and build an augmented dataset with person attribute search. Prepare inputs, configure and monitor the Kubernetes workflow, retrieve outputs, and check for missing attributes or skipped images.
Find driving clips similar to an example video and generate rain, snow, or daylight variations. Prepare run settings, monitor the Airflow workflow, and inspect retrieved clips, generated videos, and validation results.
Create synthetic defect images for detector training using clean images and masks that mark defect regions. Guide mask placement, Cosmos fine-tuning, image generation, evaluation, quality refinement, and automatic assignment of provisional labels.
Generate protein structures and sequences with specified conditions using NVIDIA's hosted API or a local NIM service. Supports binder and antibody design, reward-guided search, and inspection of generated candidates, outputs, and validation results.
Helps prepare video and image datasets with Cosmos Curator through splitting, filtering, deduplication, captioning, and embeddings, numerical representations for matching. Covers pipeline configuration, execution checks, Cosmos Dataset Search ingestion and search, and compatible handoffs to TAO data mining.
Guides CDF Auto-Labeling workflows for annotating data such as videos. Routes tasks to specialized references, covering shipped examples, stage selection, prompt and configuration changes, pipeline migration, container retrieval, and output validation.
Create and validate Cosmos Data Factory YAML configurations for changing video scenes, editing images, or generating videos. Run generation through remote model APIs, configure captions and quality checks, and retry candidates within a defined limit.
Clarify an AI agent’s purpose, boundaries, and success criteria by reviewing its code and documents, then asking focused questions about intent. The resulting draft is handed to ethos-write to create an ethos.md record.
Add optional NVIDIA Parabricks GPU steps to an existing genomics workflow while retaining its CPU path. Use runtime switches to compare outputs, execution time, and cost tradeoffs before enabling GPU processing for production.
Prepare NVIDIA Parabricks commands for individual genomics tools, translate options from CPU tools, and check runtime readiness. Get guidance matched to the target release, including reference compatibility, command flags, and unresolved requirements before execution.
Ask Auto Ontology questions in plain language, then check the generated SQL, returned data, and written answer. Supports MCP or authenticated REST access to verify that reported counts and other findings match the intended question.
Guides the definition and publication of business terms, relationships, and measures in Auto Ontology. Covers SQL validation, model import and export, and persisted-change checks to verify that saved definitions reflect the intended business meaning.