Shape agent capabilities into your workflow.

Design Atlas turns a product brief into a screen-by-screen research report. It interviews for product and brand alignment, enumerates states such as empty, loading, error, and populated, queries Mobbin for each screen, and bundles curated references into a self-contained HTML atlas. SaaS and iOS work are documented as proven; marketing sites remain experimental, and full access requires an authenticated paid Mobbin plan.

This paired workflow turns a one-line request into a Codex-ready brief with goals, scope, constraints, acceptance criteria, deliverables, validation, and output rules. A second skill on the Codex side consumes that brief, executes the task, and reports a summary, changed files, validation results, and follow-ups. It is useful when planning and implementation should be separated; the two skills are installed in different environments.

Harden Node.js and Python dependency installs with release-age cooldowns, exact-version pinning, lifecycle-script restrictions, lockfile regeneration, and trust-policy checks. The skill audits current settings, can write machine- and project-level guards, and supports a dry-run or non-interactive CLI mode. It can also generate a Socket Basics pull-request workflow, which requires a Socket API key.

This skill helps Vietnamese web developers move from understanding English to using it in real-time work situations. It offers plans, lessons, short drills, writing review, mock interviews, inline correction, and progress tracking built around reusable phrases rather than isolated vocabulary. Examples cover PHP/Laravel, React/Next.js, and Node.js/TypeScript, with progress kept in one file between sessions.

Give the skill a git repository and it turns the commit history into a single static kanban.html for PMs, QA, or business owners. The page groups 8–15 user-journey regression tasks, includes pass criteria and result capture, and keeps the full commit log with searchable impact tags and filters. It works offline without a build step, but it is an executable checklist rather than an automated test runner.

Use this language-style skill to guide AI writing toward Chinese plain description (白描): concrete, perceptible details carry the scene while explicit emotion, judgment, and causal explanation are held back. It defines requirements, mechanisms, boundaries, distinctions from nearby styles, and a seven-point self-check. The skill is application-independent, so it can support fiction, screenplays, RPG dialogue, journalism, or other prose.

This extension applies Chinese plain-description principles to CRPG and choose-your-own-adventure dialogue nodes. It expects a structured input containing known player information, the foreground goal, background seeds, and one of five information types, then checks preconditions and node-specific rules for narration, NPC speech, player narration, or options. Load the general baimiao skill first; this layer adds the game-specific contract and regression checks.

Use this Claude Code plugin to scaffold or extend Expo React Native apps on a fixed TypeScript architecture. Its agents cover app setup, feature work, standards review, simulator or end-to-end runs, and EAS build, update, and submission; feature work must pass review before and after implementation. The locked stack includes Redux Toolkit, RTK Query, React Navigation, react-hook-form with Yup, i18n, and Maestro, so it suits teams willing to adopt those choices.

Run an independent review cycle on a written plan before implementation. The skill dispatches a domain-specific reviewer for software, Brazilian Portuguese judicial drafting, or a generic case, walks through findings, and uses a deterministic Python validator that fails while blocking findings remain open. It is useful for teams that want explicit acceptance and validation before execution; the judicial mode is tailored to CNJ Resolution 615/2025 audit obligations.

Use this Claude skill to structure EU and cross-border VAT questions in French, Dutch, English, or German. It helps qualify B2B and B2C transactions, assess place of supply, OSS/IOSS, VIES, SaaS, Brexit and Northern Ireland cases, and prepare invoice considerations. It deliberately avoids storing volatile rates, thresholds, and dates, requiring web verification for current figures; its output is general analysis, not tax advice.

This Claude skill covers Belgian accounting workflows in French and Dutch, including PCMN/MAR journal entries, balance sheets, income statements, BE GAAP, annual accounts, ASBL/VZW reporting, and sole-trader cases. It is an educational productivity aid, not professional accounting, tax, or legal advice, and draft figures still need verification.

This Claude skill writes Spanish-language X posts and threads, offering variants and rationale, and can plan weekly calendars in Markdown or Excel. It can research a reference account and score drafts with signals from xai-org/x-algorithm; browser-based posting requires an authenticated Chrome session.

This skill scaffolds an Apache-2.0 project with a license, ignore rules, README, CI, repository topics, an initial commit, and GitHub and Codeberg remotes. It can push to both hosts and follows preset CryptoJones and Nebraska conventions, so users should check those project-specific defaults before running it.

This skill coordinates a pull request lifecycle across GitHub and Codeberg: branch, commit, push, open issues and PRs on both hosts, squash-merge, reconcile divergent hashes, and remove branches. It expects gh authentication, a CB_TOKEN environment variable, matching remotes, and preset CryptoJones, Aaron, and Nebraska conventions.

This skill wraps the upstream chbs command instead of reimplementing passphrase generation. It produces xkcd-936-style six-word phrases, shows an entropy caption, and supports count, capitalization, and separator variants; the upstream package must be installed and chbs must be available on PATH.

This skill wraps the upstream interruptingcow package and its moo command to notify an operator through Telegram when a long task finishes, fails, or crosses a time threshold. It does not reimplement the HTTP client; users must provide a Telegram bot token and chat ID, and handle the underlying task lifecycle themselves.

This skill manually reprioritizes the Claude Code task queue by deleting and recreating tasks. Its planned triage CLI only suggests an order from task signals, while a human confirms the change; it never executes queue reordering automatically.

This skill routes research through nine scenarios or a generic mode, covering question framing, source collection, conflict tracking, four critique passes, citation validation, and gap logging. It produces a research-vault with query, plan, sources, findings, conflicts, gap-log, critique, and report files, while separating facts, inferences, and recommendations.