Progressive-disclosure instructions an agent loads on demand. Cheap until invoked, which is why the per-session figure on each card is usually two digits.
Mandatory writing quality rules that eliminate AI-writing patterns from ALL generated output. This skill applies passively to every response (prose, emails, code comments, creative writing) shaping generation from the start. It also provides an active review pass when invoked explicitly. ALWAYS trigger this skill. It…
Prioritize features, tasks, bugs, initiatives, or any backlog of items using the VERDICT framework. Use this skill whenever the user wants to decide what to build first, triage a backlog, sequence a roadmap, compare multiple initiatives, or score items against a consistent rubric. Also trigger when the user says…
Comprehensive SOP for Amazon SageMaker AI — SDK v3 code generation, serverless model customization (SFT, DPO, RLVR, RLAIF), LLM inference endpoints, model training (classical ML + LLM fine-tuning), HyperPod clusters (training + inference), Model Monitor, and AutoML with AutoGluon. Generates correct, working code every…
Full feature development lifecycle — PRD, critical review, task breakdown, implementation, and quality gates. Use when the user provides a feature description and wants to go all the way from idea to working, reviewed, tested code.
Upload images to img402.dev for embedding in GitHub PRs, issues, and comments. Images 1MB or under are free AND permanent — no payment, no auth — which covers virtually every screenshot. Larger images (up to 10MB) are free for 30 days, or permanent for $0.01 USDC via x402. Use when the agent needs to share an image in…
Read an article and produce a paste-and-go gpt-image-2 prompt. Two cover families — anime-pop (EVA / Persona 5 / Akira) and mega-chinese-type (giant Chinese perspective typography, esports key visual, editorial sports, movie poster). Outputs decision JSON + English image prompt. Use when user wants to generate an…
Records decisions and documentation. Use when making architectural decisions, changing public APIs, shipping features, or when you need to record context that future engineers and agents will need to understand the codebase.
An automated testing setup for web applications that uses Playwright to control a real browser and checks complete user journeys with the live backend.
Run a comprehensive code audit on any codebase. Analyzes security vulnerabilities, hardcoded secrets, dependency CVEs, test coverage, code structure, and AI-generated code patterns. Generates CODEAUDITREPORT.md with findings, severity ratings, and remediation guidance. Intelligently selects and orchestrates the best…
Generates comprehensive test suites for code including unit tests, integration tests, and end-to-end tests. Use when creating tests for functions, classes, APIs, or entire modules, covering edge cases, error conditions, and ensuring high code coverage.
A placeholder skill description written in Chinese that says a skill should explain its purpose and when it should be triggered. It provides a rough template with sections for inputs, workflow, output, and references.
Controls response length and pacing in all conversations. Use this skill in EVERY chat by default. It keeps replies short, bans preamble and recap, and forces one-thing-at-a-time collaboration instead of dumping full plans. Especially important during new-project brainstorming, where the natural failure mode is…
Run a Claude–Codex adversarial review of a plan or code change. Use only when the user explicitly asks for a debate, cross-review, or second opinion; every round spends a paid turn on the other provider.
Scans files changed in the current branch for security vulnerabilities including OWASP Top 10, hardcoded secrets, injection flaws, and insecure patterns. Use when the user wants a security audit, vulnerability check, or security review of their changes.
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
Generates a portable, single-file HTML artifact from free-form planning output. Takes any input — issue body, RFC, spike summary, grilling transcript, inline content, file path, Google Doc content, or braindump — and produces either a static report or an animated slideshow, both styled with Elastic's EUI Borealis…
A setup guide for starting a Go project in the current directory. Go is a compiled programming language, and the guide checks existing source files and project configuration before continuing.
Use when the user wants to read, catch up on, send, reply to, react to, or @mention in Microsoft Teams chat messages ("check Teams", "any messages in ", "reply in ", "message the chat", "react to that message", "@mention someone", "catch me up on Teams"). Drives the new Teams web (teams.cloud.microsoft) in Chrome via…
Turn a feature specification or implementation plan into a standalone local HTML report that helps a person understand the intended architecture, flows, UI states, contracts, scope, and decisions before implementation. Use when a plan is substantial enough to benefit from visual review. Do not use for recapping code…
★not rated 2 1mo agoA65 tokens
originalMIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: