hungson175

Headless browser automation CLI for AI agents. Use this skill when you need to automate web interactions, scrape web pages, test web applications, or perform any browser-based tasks. Triggers include "browse", "automate browser", "scrape website", "test web app", "fill form", "take screenshot", or any web automation request. Provides accessibility-first element targeting with semantic locators.

Toolkit for testing Android applications using Appium with Espresso driver. Supports automated UI testing, gesture interactions, element discovery, and screenshot capture. Use when testing native Android apps (.apk files), verifying mobile UI functionality, automating gestures (swipe, scroll, tap), or running blackbox tests on Android devices/emulators.

This skill should be used when users want to create, configure, or debug Claude Code hooks. Hooks are shell commands that execute at various points in Claude Code's lifecycle (PreToolUse, PostToolUse, Stop, SessionStart, etc.). Use this skill when users ask to create custom automation, file protection, code formatting hooks, notifications, or any lifecycle-based triggers for Claude Code.

Retrieve universal coding patterns from vector database. Auto-invokes before complex tasks or when user says "--recall". Searches relevant role collections based on task context.

Store universal coding patterns into vector database. Auto-invokes after difficult tasks with broadly-applicable lessons. Trigger with "--store" or when user expresses frustration (strong learning signals).

Create custom slash commands for Claude Code. Use this skill when the user asks to "create a command", "add a slash command", "make a custom command", "write a command for...", or needs help with command frontmatter, arguments ($ARGUMENTS, $1, $2), file references (@syntax), bash execution in commands, or organizing commands in project/personal/plugin locations.

Find reputable GitHub repositories with quality filters (stars, recency, license). Use this skill when the user needs to find code samples, reference implementations, libraries, or templates on GitHub. Triggers on requests like "find GitHub repos for X", "search GitHub for Y examples", "find a good library for Z", or any request to discover quality open-source code. Requires gh CLI (version 2.14+).

Toolkit for testing hand drawing applications including signature capture, canvas drawing, and art apps. Supports touch gesture simulation for drawing strokes, image comparison for output verification, and multi-stroke automation. Use when testing drawing/canvas features, signature capture, annotation tools, or any app requiring touch path validation.

Retrieve coding patterns from GLOBAL vector database (cross-project learning). Auto-invokes when TodoWrite has >3 tasks or when user says "--recall". Searches relevant role collections based on task context.

Store coding patterns into GLOBAL vector database (cross-project learning). Auto-invokes after difficult tasks with broadly-applicable lessons. Trigger with "--store" or when user expresses frustration (fuck, bitch, dog, bad words = strong signal to call memory-store).

Retrieve universal coding patterns from vector database. Auto-invokes before complex tasks or when user says "--recall". Searches relevant role collections based on task context.

Store universal coding patterns into vector database. Auto-invokes after difficult tasks with broadly-applicable lessons. Trigger with "--store" or when user expresses frustration (strong learning signals).

Use when the user needs Vietnam (VN) financial-market data in Python — daily/historical stock prices (OHLCV), company fundamentals (income/balance/cashflow/ratios), mutual-fund NAV and holdings, market indices and constituents (VNINDEX, VN30, HNX, UPCOM), domestic and world gold prices, foreign-exchange rates vs VND, major crypto OHLCV, or cross-country macro indicators (GDP, CPI, inflation, unemployment). Provides the clean-room, no-API-key `vnfin` library with multi-source failover and typed, unit-explicit results. Trigger on requests like "get FPT stock prices", "Vietnam GDP", "VNINDEX history", "SJC/gold price", "USD to VND", "fund NAV", "VN30 constituents", or any task building a VN-market data/analysis/advisor tool.

Convert a native-text PDF book/manuscript into a clean, faithful EPUB (headings, paragraphs, dialogue, footnotes, bordered tables, embedded photos) ready for Kindle. Use when Boss wants to read a PDF book on Kindle/e-reader, or asks to "convert this PDF to epub", "turn this book into an ebook", or names this skill directly. Handles books with ruling-line tables (lookup grids, reference charts) as well as continuous prose -- NOT for business documents/decks with charts/graphics-as-data, use document-processor for those.

Ingest business documents (.pptx, .pdf, .xlsx, .docx) into FAITHFUL markdown for a RAG knowledge base or review — VLM-first (Vision-Language Model), NOT OCR. Use this whenever the user wants to convert, ingest, extract, or "turn into markdown/text" any slide deck, report, spreadsheet dashboard, or PDF — ESPECIALLY chart-, dashboard-, or image-heavy and Vietnamese/mixed-language documents where chart numbers, tables, and verbatim text must survive. Trigger on phrases like "convert this deck/report/pptx/pdf to markdown", "ingest these documents", "extract the data/tables/charts from this presentation", "build a knowledge base from these files", or any document→markdown task where faithfulness matters. Prefer this over a plain text-extraction/OCR approach, which silently drops charts and merged-cell layouts. Validated 10/10 on real documents by an independent Opus-4.8 faithfulness judge (see references/validation.md).

Create ONE self-contained personal AI agent in its own folder, with a two-part brain: file-based memory plus a knowledge base built with Andrej Karpathy's "LLM wiki" method (append-only raw/ + distilled wiki/). The agent is standalone — it does NOT report to or coordinate with any other agent. Use this whenever someone asks to "create an agent", "make/spawn/bootstrap a new agent", "set up a personal assistant/tutor/research agent", "scaffold an agent with memory and a knowledge base", or wants a dedicated agent for one topic, project, or domain. Do NOT use for multi-agent teams.

Mid-depth research orchestrator sitting between WebSearch and external deep-research tools (Gemini/ChatGPT). Spawns 2-4 parallel research-assistant sub-agents, synthesizes findings, writes a cited Markdown report to docs/research/. Use for research writeups, multi-source comparisons, options/candidate ranking, or any topic exploration deeper than a single WebSearch. The calling agent decides when to invoke based on user intent.

Prompting best practices for xAI Grok-4-Fast non-reasoning model (grok-4-fast-non-reasoning). Use this skill when writing or optimizing system prompts, tool docstrings, or instructions for the gieoque app's LLM backend. Triggers on: write system prompt, optimize prompt, grok prompt, improve tool description, grok-4-fast.