Eunkwang Shin
Community@gracefullight · Ultimo, NSW
I literally can’t sleep if I don’t code for even a single day.
Agent Skills by Eunkwang Shin
Showing 65 vetted skills indexed across 5 GitHub repositories.
gracefullighttensorflow-neural-networks
Build and train neural networks using TensorFlow Keras APIs and custom layer implementations.
gracefullighttensorflow-data-pipelines
Build tf.data pipelines for dataset loading, transformation, batching, and prefetching in TensorFlow.
gracefullightshap
Compute SHAP values and generate visualizations to explain machine learning model predictions.
gracefullightx-research
Searches X/Twitter via API to research topics, threads, and accounts with sourced briefings.
gracefullightdask
Scale pandas and NumPy workflows to larger-than-memory datasets with parallel and distributed computing.
gracefullightget-available-resources
Detect CPU, GPU, memory, and disk resources and generate computational strategy recommendations.
gracefullightexploratory-data-analysis
Analyze scientific data files across 200+ formats and generate markdown EDA reports.
gracefullightseaborn
Create statistical visualizations from pandas DataFrames using seaborn plotting functions.
gracefullighttensorflow-model-deployment
Export, quantize, and serve TensorFlow models via SavedModel, TFLite, and TensorFlow Serving.
gracefullightplotly
Create interactive Python charts with hover, zoom, and export to HTML or static images.
gracefullightmatplotlib
Create static, animated, and interactive plots with matplotlib's pyplot and object-oriented APIs.
gracefullightscientific-visualization
Create publication-ready scientific figures with matplotlib, seaborn, and plotly for journal submission.
gracefullightsagemaker-endpoint-deployer
Automate AWS SageMaker endpoint deployment across setup, deployment, and validation stages.
gracefullightscikit-learn
Build, evaluate, and tune machine learning models in Python with scikit-learn.
gracefullightoma-observability
Route observability work into setup, investigation, alerting, tracing, tuning, and routing playbooks.
gracefullightoma-orchestrator
Decompose complex feature requests into priority-tiered tasks and dispatch specialist agents in parallel.
gracefullightoma-translator
Translate UI strings, documentation, and marketing copy while preserving placeholders and structure.
gracefullightoma-qa
Performs evidence-based quality assurance reviews for security, performance, accessibility, and correctness.
gracefullightoma-recap
Generate themed Markdown work summaries from AI conversation histories.
gracefullightoma-scm
Guide safe Git configuration management with staging and governance checks.
gracefullightoma-pm
Decompose product requirements into dependency-aware JSON planning artifacts with acceptance criteria.
gracefullightoma-mobile
Implement and verify cross-platform mobile features for Flutter or React Native.
gracefullightoma-scholar
Generate and validate Knows v0.9.0 `.knows.yaml` research-paper sidecars.
gracefullightoma-image
Generate AI images via authenticated Codex, Pollinations, and Gemini providers.
Frequently Asked Questions About Eunkwang Shin
FAQPage SchemaWhat tasks can I accomplish with gracefullight's skills?▼
You can build and deploy TensorFlow neural networks, explain models with SHAP, run scikit-learn pipelines, perform EDA on 200+ scientific formats, create publication-quality figures with matplotlib/seaborn/plotly, orchestrate multi-agent development, manage Terraform infrastructure, and backtest trading strategies with Yahoo Finance data.
Who is the target audience for these skills?▼
Data scientists, ML engineers, and scientific researchers benefit from the TensorFlow, SHAP, Dask, and visualization skills, while software teams use the OMA orchestration, QA review, Git SCM, and architecture ADR skills. Quantitative analysts use the stock-analysis and backtesting skills.
How do the machine learning skills work together in practice?▼
Start with get-available-resources to detect CPU/GPU/memory, run exploratory-data-analysis on input files, build pipelines with tensorflow-data-pipelines or scikit-learn, train models, interpret them with SHAP, and deploy via tensorflow-model-deployment or the SageMaker endpoint deployer.
Are gracefullight's skills open source and free to use?▼
Yes. The natively documented skills carry permissive licenses: SHAP, plotly, scientific-visualization, exploratory-data-analysis, and get-available-resources use MIT, while Dask, seaborn, and scikit-learn use BSD-3-Clause. Matplotlib uses its own PSF-based license.
What prerequisites or dependencies do these skills require?▼
TensorFlow skills declare Bash and Read tool access. X-research requires a pay-per-use X API plan with recent search limited to the last 7 days. OMA-image needs authenticated Codex, Pollinations, or Gemini providers, and orchestrate relies on Gemini CLI with MCP Memory coordination.