Skills

planning

Discovers user intent and generates a structured, step-by-step plan for model customization workflows. This skill must always be activated alongside any other skill when the user's request relates to model customization — including fine-tuning, training, building, customizing, reviewing data, or get

datahub-search

Use this skill when the user wants to search the DataHub catalog, discover entities, answer ad-hoc questions about their data, find datasets, or browse by platform or domain. Triggers on: "search DataHub", "find datasets", "who owns X", "what tables contain PII", "what columns does X have", or any r

bedrock

Deep-dive into Amazon Bedrock — model selection, agents, knowledge bases, guardrails, prompt engineering, and cost modeling. This skill should be used when the user asks to "build with Bedrock", "select a Bedrock model", "design a Bedrock agent", "set up a knowledge base", "configure guardrails", "e

together-fine-tuning

LoRA, full fine-tuning, DPO preference tuning, VLM training, function-calling tuning, reasoning tuning, and BYOM uploads on Together AI. Reach for it whenever the user wants to adapt a model on custom data rather than only run inference, evaluate outputs, or host an existing model.

model-evaluation

Generates python code that evaluates SageMaker models. Supports two evaluation types: LLM-as-Judge and Custom Scorer. Use when the user says "evaluate my model", "run a benchmark", "test model performance", "how did my model perform", "compare models", or other similar requests.

pixeltable

Build multimodal AI applications with Pixeltable — declarative tables replace LangChain + pandas + vector DB with one system. Automates chunking, embedding, retrieval, tool-calling agents, and 25+ AI provider integrations (OpenAI, Anthropic, Gemini, etc.) via computed columns that run on insert. Use

output-credentials-env-vars

Wire encrypted credentials to environment variables using the credential: convention. Use when setting up LLM provider keys (ANTHROPIC_API_KEY, OPENAI_API_KEY) or any env var that should come from encrypted credentials.

read-file

Read any data file (CSV, JSON, Parquet, Avro, Excel, spatial, SQLite) or remote URL (S3, HTTPS). Use when user references a data file, asks "what's in this file", or wants to preview/profile a dataset. Not for source code.

plagiarism-checker

Scans lyrics for phrases that may match existing songs using web search and LLM knowledge. Use before release to check for unintentional borrowing.

data-journalism

Data journalism workflows for analysis, visualization, and storytelling. Use when analyzing datasets, creating charts and maps, cleaning messy data, calculating statistics or building data-driven stories. Essential for reporters, newsrooms and researchers working with quantitative information.

full-output-enforcement

Overrides default LLM truncation behavior. Enforces complete code generation, bans placeholder patterns, and handles token-limit splits cleanly. Apply to any task requiring exhaustive, unabridged output.

hf-cloud-serving-image-selection

Pick the right serving container for a SageMaker model deployment and find its current image URI. Use this skill whenever about to deploy a model to a SageMaker endpoint and an image URI needs to be chosen — including when the user says "deploy this LLM", "host this HuggingFace model", "serve this f

fiftyone-dataset-curation

End-to-end dataset curation for FiftyOne: inspect schema and quality, audit annotations, analyze class distributions, explore embeddings, find duplicates, create curated subsets, and build train/val/test splits. Works with any computer vision dataset type.

notebook-guidance

This skill guides the use of Jupyter notebooks for data analysis, exploration, and visualization, particularly with BigQuery. It outlines best practices for notebook execution and validation (supporting both cell-by-cell execution and full notebook generation depending on tool availability), library

firecrawl-scrape

Extract clean markdown from any URL, including JavaScript-rendered SPAs. Use this skill whenever the user provides a URL and wants its content, says "scrape", "grab", "fetch", "pull", "get the page", "extract from this URL", or "read this webpage". Handles JS-rendered pages, multiple concurrent URLs

huggingface-llm-trainer

Train or fine-tune language and vision models using TRL (Transformer Reinforcement Learning) or Unsloth with Hugging Face Jobs infrastructure. Covers SFT, DPO, GRPO and reward modeling training methods, plus GGUF conversion for local deployment. Includes guidance on the TRL Jobs package, UV scripts