Copy this prompt into the AI assistant you're using.
Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/balintdecsi/skills.Loading the repository README…
Showing 20 of 20 collected skills.
Visual direction and build process for new or reshaped web UI — grounding the design in its subject, a compact token plan (palette, type, layout, principles), avoiding templated AI-generated defaults, restrained motion, interface copy, and a self-critique…
Review web UI source code — HTML, CSS, JSX/TSX, and Vue, Svelte, or Astro templates — against the Vercel Web Interface Guidelines, tiered by this repository's overrides into required, recommended, and style findings in terse file:line form. Covers…
Query and apply The Website Specification — a platform-agnostic contract for what a good website does, with items tagged required, recommended, optional, or avoid and backed by WHATWG, W3C, IETF, IANA, and WCAG sources. Use when deciding what a site should…
Technical setup skill for analytics and data science projects — repository scaffolding, mandatory dev/prod and bronze/silver/gold structure, numbered notebooks, uv environments, dotenv, notebook-output hooks, branching and Conventional Commits, .gitignore…
Helps users discover and, with permission, install agent skills when they ask to find a skill, ask whether a skill exists, or want to extend agent capabilities. Check installed and repository-local skills first; search the external ecosystem only when no…
Best-practice suggestions for geospatial data science in Python — vector data with GeoPandas/Shapely, raster data with rasterio/rioxarray/xarray, OpenStreetMap via OSMnx, geocoding, H3 spatial indexing, basemaps with contextily, and spatial statistics or…
Operate Maritime (https://maritime.sh) from the maritime CLI: authenticate, create and deploy agents, set secrets, chat, schedule wakes, and debug with logs and exec. Use when the user mentions Maritime, maritime.sh, the maritime command, hosting an agent in…
Best-practice suggestions for building, comparing, or reviewing predictive supervised ML models in Python with scikit-learn — Pipelines, ColumnTransformer, cross-validation, hyperparameter search, leakage prevention, candidate leaderboards, one-time held-out…
Guide for using uv, the Python package and project manager. Use for Python projects, scripts, packages, interpreters, virtual environments, dependency locking, and one-off Python tools, especially when uv.lock or uv-generated requirements files are present.…
Write a Conventional Commits message compressed to intent only. Use for "write a commit", "commit message", /commit or /caveman-commit.
Tool-assisted code review using the best review, quality, coverage, and analysis tools available, such as CodeScene, Codecov, or CodeRabbit, plus direct code inspection. Use for any explicit code, diff, commit, or PR review and when implementation work calls…
Best practices for designing and reviewing data warehouses and analytical pipelines using the mandatory bronze/silver/gold medallion architecture, validations-as-code, idempotent transforms, explicit table grain and keys, and in-pipeline data-quality checks.…
Suggestions for drafting, reviewing, or critiquing an Analytics Project Brief — the pre-code artifact that defines the decision, metrics, counter-metrics, stakeholders, methodology, scope, success criteria, risks, and pre-mortem. Use to scope an analytics…
Enforces authenticated gh CLI workflows over unauthenticated curl, wget, or WebFetch patterns. Use for GitHub repositories, URLs, API access, raw files, pull requests, issues, releases, or private content. Do not use for non-GitHub services or when plain…
Read and edit Joplin notes from the terminal with the Joplin CLI (`joplin`), and keep the local copy safely in sync with the sync target. Use when the user asks to read, search, add to, or update a Joplin note or notebook, run `joplin sync`, or fix a…
Practical compliance guidance for software and data projects — EU data protection, GDPR lawful basis, consent records, minors, retention, data-subject rights, access control, ePrivacy/cookies, AI Act considerations, privacy pages, records of processing, and…
Use the Notion CLI (`ntn`) to interact with the Notion API, manage workers, and upload files. Use when the user asks to "call the Notion API", "deploy a worker", "upload a file to Notion", "create a page", "query a database", or any task involving the `ntn`…
Suggestions for the earliest stage of a revenue-generating product — problem and idea selection, hypotheses, Mom Test interviews, personas, beachhead markets, TAM, competitors, riskiest-assumption tests, MVP experiments, business models, pricing, and…
Best-practice suggestions for building, interpreting, or reviewing statistical and inferential models in Python — OLS and logistic regression with statsmodels, robust standard errors, regression tables, confidence and prediction intervals, hypothesis tests,…
Best-practice suggestions for analyzing, building, comparing, or reviewing forecasts for dated or ordered data such as demand, energy, sales, traffic, sensors, macro, or finance series. Covers datetime indexing, resampling, trend and seasonality, lag…
The repository README could not be loaded. You can still explore its Skills below.