🎯Skills49
Skill
Skill
A skill that lets AI agents control the user's real browser through a local daemon, supporting navigation, clicking, typing, form filling, screenshots, PDF export, and network monitoring while using the user's actual login sessions.
Browser automation CLI for AI agents — website interaction, form filling, screenshots, data extraction, web app testing, and Electron app automation (VS Code, Slack, Discord, Figma).
A skill for using Exa search tools via MCP, providing web search, code/documentation search, and company research capabilities with configurable parameters for result count, content length, and search type.
An agent skill that provides a CLI and MCP server for searching and retrieving SVG icons from 200+ libraries via Iconify, with support for batch downloads, project syncing, and smart recommendations.
Fetches current third-party API and library documentation via the chub CLI before writing code, ensuring accurate and up-to-date API reference instead of relying on training data. Supports searching, retrieving, and annotating docs.
Helps build AI chat interfaces using the ai-elements component library, which provides pre-built conversations, messages, tool displays, and prompt inputs built on shadcn/ui.
Systematically explores and tests a web application using browser automation to find bugs, UX issues, and other problems. Produces a structured report with full reproduction evidence including step-by-step screenshots, repro videos, and detailed reproduction steps for every issue.
A skill for professional DOCX document creation, editing, and formatting using OpenXML SDK (.NET). Supports three pipelines: creating new documents from scratch, filling/editing existing documents, and applying template formatting with XSD validation.
Guides AI agents to build visually premium landing pages, apps, and prototypes with art-direction-first principles: restrained composition, image-led hierarchy, strong typography, and deliberate motion design rather than generic card-based layouts.
A skill for creating, reading, analyzing, editing, and validating Excel spreadsheet files (.xlsx, .xlsm, .csv, .tsv) using XML direct-edit techniques, with support for formula management, professional financial formatting, and zero-format-loss editing.
A skill for generating, editing, and reading PowerPoint presentations. Supports creating decks from scratch with PptxGenJS (cover, TOC, content, section divider, and summary slides), editing existing PPTX files via XML workflows, and extracting text with markitdown.
A skill that debugs and fixes failing GitHub PR checks by using the gh CLI to inspect GitHub Actions logs, summarize failure context, draft a fix plan, and implement changes only after explicit user approval.
A Claude Code skill for automating Electron desktop applications (VS Code, Slack, Discord, Figma, Notion, Spotify, etc.) via the Chrome DevTools Protocol through agent-browser, enabling natural language control and testing of any running Electron app.
A skill that helps address review and issue comments on open GitHub pull requests for the current branch using the gh CLI. Fetches all comments, presents numbered summaries, and applies fixes for user-selected items.
A PDF generation skill that handles three workflows: creating new PDFs from scratch with 15+ design templates, filling form fields in existing PDFs, and reformatting documents with professional styling and print-ready output.
An ultra-compressed communication mode skill that reduces token usage by approximately 75% while maintaining full technical accuracy. Supports multiple intensity levels from lite to ultra, plus classical Chinese (wenyan) modes.
A comprehensive GLSL shader skill covering 36 techniques for real-time visual effects including ray marching, SDF modeling, fluid simulation, particle systems, procedural generation, lighting, and post-processing. Each technique includes implementation guides and code templates compatible with ShaderToy.
A documentation lookup skill that uses the Context7 CLI to retrieve current documentation, API references, and code examples for any developer library or framework, ensuring answers reflect the latest API changes rather than stale training data.
A debugging skill that enforces systematic root cause investigation before attempting any fixes. Guides through four phases of analysis including error reading, consistent reproduction, recent change review, and diagnostic instrumentation for multi-component systems.
Fetches dependency source code so AI agents can read library implementations, not just types. Supports npm, PyPI, and crates.io packages with automatic version detection from lockfiles and global caching at ~/.opensrc/.
A test-driven development skill that enforces red-green-refactor workflow before writing implementation code, emphasizing behavior verification through public interfaces rather than implementation-coupled tests.
An autonomous goal-directed iteration engine inspired by Karpathy's autoresearch that applies a constraint-driven Modify/Verify/Keep/Discard cycle to any work, with built-in safety guardrails including atomic commits and mandatory verification.
Attaches a read-only Chrome DevTools Protocol client to a running browser session to capture full CDP event streams, screenshots, and DOM snapshots. Bisects the trace into per-page searchable buckets for debugging browser automation failures.
An AI skill that surfaces architectural friction and proposes deepening opportunities in codebases. Identifies shallow modules, pass-through layers, and coupling issues to improve testability and AI-navigability using a consistent vocabulary of depth, seams, and adapters.
A skill that migrates test files from TypeScript `as` type assertions to @total-typescript/shoehorn, enabling type-safe partial test data with `fromPartial()` and `fromAny()` helpers.
An interactive grilling skill that challenges your development plan against the existing domain model, sharpens terminology, and updates documentation such as CONTEXT.md and ADRs inline as decisions are made.
Behavioral guidelines for AI coding agents derived from Andrej Karpathy's observations on common LLM coding mistakes. Promotes surgical changes, explicit assumptions, and verifiable success criteria to reduce overcomplication.
An interview skill that stress-tests your plan or design by asking relentless, targeted questions. It walks through each branch of the decision tree one at a time, providing recommended answers and exploring the codebase when possible.
A disciplined diagnosis loop for hard bugs and performance regressions, following a structured process: reproduce, minimise, hypothesise, instrument, fix, and regression-test.
A one-command skill that stages all changes, commits, pushes, and opens a GitHub pull request using the GitHub CLI (gh). Handles branch creation, naming conventions, status checks, and generates detailed PR descriptions covering the issue, root cause, and fix.
A collection of 40 reusable AI coding agent skills compatible with Codex, Claude Code, Cursor, and other tools, covering docs lookup, browser automation, GitHub Actions debugging, prompt engineering, planning workflows, frontend design, and office document authoring.
Part of OK Skills, a curated collection of 40 reusable AI coding agent skills and SKILL.md playbooks for Claude Code, Cursor, Codex, and other agents, covering research, docs lookup, browser automation, GitHub workflow, planning, and frontend design.
A curated collection of 40 reusable AI coding agent skills and playbooks for Codex, Claude Code, Cursor, and other tools, covering docs lookup, browser automation, GitHub Actions, prompt engineering, frontend design, and document authoring.
A curated collection of 40 reusable AI coding agent skills and playbooks for Codex, Claude Code, Cursor, and OpenClaw, covering docs lookup, browser automation, GitHub Actions debugging, planning workflows, and PDF/Office document authoring.
A curated collection of 40 reusable AI coding agent skills and playbooks for Claude Code, Codex, Cursor, and other SKILL.md-compatible tools. Includes 32 top-level skills plus 8 vendored GSAP animation skills.
A collection of 40 reusable AI coding agent skills for Codex, Claude Code, Cursor, and other SKILL.md-compatible tools, covering docs lookup, browser automation, GitHub Actions debugging, prompt engineering, and planning workflows.
A collaborative design skill that helps turn rough ideas into fully formed designs and specifications through structured dialogue, exploring user intent, requirements, and design before any implementation begins.
Part of the OK Skills collection, a curated set of 40+ reusable AI coding agent skills compatible with Codex, Claude Code, Cursor, and other SKILL.md-compatible tools, covering docs lookup, browser automation, planning workflows, and frontend design.
Skill
Skill
Skill
Skill
Skill
Skill
Skill
Skill
Skill