Agent Skills by feba
Showing 66 vetted skills indexed across 1 GitHub repositories.
html-plan
Generate structured HTML plans with CSS variables and dark mode.
domain-modeling
Create and manage domain models with terminology, ubiquitous language, and architectural decisions.
using-git-worktrees
Create isolated Git worktrees and set up Node.js, Rust, Python, or Go projects.
agent-browser
Automate web interactions including navigation, form filling, and data extraction.
setup-pre-commit
Automate Git pre-commit hook setup with Husky, lint-staged, and linters.
figma-generate-design
Automate Figma screen creation and updates using design system tokens.
test-driven-development
Apply test-first development with the red-green-refactor cycle.
xinchao-work-assistant
Manage local AI work directories and track Markdown tasks with PowerShell.
web-prototype
Generate HTML prototypes from predefined templates and custom layouts.
systematic-debugging
Apply a four-phase debugging process to identify root causes of technical issues.
setup-matt-pocock-skills
Automate engineering skill setup for issue trackers, triage labels, and domain docs.
writing-great-skills
Provide guidelines for developing predictable and efficient AI agent skills.
paper-review-skills
Automate academic paper reviews with structured frameworks and AI-assisted reading.
humanizer-zh
Remove AI writing traces from text using NLP patterns.
obsidian-vault
Manage and search Obsidian vault notes by filename and content.
hatch-pet
Create Codex-compatible animated pets and sprite sheets from character art.
handoff
Generate a handoff document summarizing conversation context and future session focus.
using-superpowers
Guide skill invocation and prioritization across Claude Code, Copilot CLI, and Gemini CLI.
dispatching-parallel-agents
Dispatch parallel agents to execute independent tasks concurrently.
diagnosing-bugs
Diagnose complex bugs and performance regressions through a structured feedback loop.
git-guardrails-claude-code
Block dangerous git commands before execution in Claude Code environments.
deep-research
Automate systematic academic literature reviews across search, analysis, and report compilation.
ask-matt
Route users to appropriate skills or flows within a repository.
improve-codebase-architecture
Analyze codebase architecture and generate an HTML report of refactoring opportunities.