feiskyer-claude-code-settings
🏪Marketplacefeiskyer/claude-code-settings
|Claude Code settings and skills for spec-driven development workflows
Overview
feiskyer/claude-code-settings is a curated collection of Claude Code settings, custom commands, skills, and subagents designed for spec-driven development workflows. It includes a LiteLLM proxy configuration for using GitHub Copilot as an LLM gateway, along with specialized skills for autonomous task execution, image generation, YouTube transcription, and more.
Key Features
- Spec-Driven Workflow — Commands and skills organized around specification-first feature development with structured planning and execution
- LiteLLM Proxy Integration — Configures LiteLLM Proxy Server as an LLM gateway to GitHub Copilot, enabling cost-effective model access
- Multiple Installation Methods — Install as a Claude Code plugin marketplace, use
npx skillsfor skills-only installation, or manually clone to~/.claude - Specialized Skills — Includes autonomous-skill for long-running tasks, nanobanana-skill for image generation, youtube-transcribe-skill for transcript extraction, and kiro/spec-kit skills for workflow management
- Individual Skill Installation — Each skill (codex, autonomous, nanobanana, kiro, spec-kit, youtube-transcribe) can be installed independently
Who is this for?
Claude Code users who want an opinionated, productivity-focused configuration with spec-driven development workflows. It is especially suited for developers who want to leverage GitHub Copilot through LiteLLM proxy and need specialized skills for autonomous task execution and content creation.
Add this Marketplace
/plugin marketplace add feiskyer/claude-code-settingsPlugins in this Marketplace
autonomous-skill
claude-code-settings
Curated collection of Claude Code settings, custom commands, skills and sub-agents for enhanced development workflows including spec-driven feature development, code analysis, GitHub integration, and knowledge management.
codex-skill
Curated collection of Claude Code settings, custom commands, skills, and sub-agents for enhanced development workflows including spec-driven feature development, code analysis, and GitHub integration.
kiro-skill
Curated collection of Claude Code settings, custom commands, skills, and sub-agents for enhanced development workflows including spec-driven feature development, code analysis, and GitHub integration.
nanobanana-skill
Claude Code plugin for AI image generation, part of a curated settings and skills collection for enhanced development workflows.
spec-kit-skill
Spec Kit skill for spec-driven development workflows from a curated Claude Code settings collection. Provides structured feature development with planning, analysis, and implementation stages.
youtube-transcribe-skill
More from this repository10
GitHub Spec-Kit integration providing a 7-phase constitution-driven workflow for feature development: Constitution, Specify, Clarify, Plan, Tasks, Analyze, and Implement, with structured file storage in the .specify/ directory.
Interactive spec-driven development workflow that transforms ideas into feature specifications using EARS format requirements, design documents with architecture and data models, and actionable task lists for test-driven incremental implementation.
Curated collection of Claude Code settings, custom commands, skills, and sub-agents for enhanced development workflows including spec-driven feature development, code analysis, and GitHub integration.
Curated collection of Claude Code settings, custom commands, skills, and sub-agents for enhanced development workflows including spec-driven feature development, code analysis, and GitHub integration.
Curated collection of Claude Code settings, custom commands, skills, and sub-agents for enhanced development workflows including spec-driven feature development, code analysis, and GitHub integration.
Curated collection of Claude Code settings, custom commands, skills, and sub-agents for enhanced development workflows including spec-driven feature development, code analysis, and GitHub integration.
Orchestrates multi-agent deep research workflows by breaking down research goals into parallel subtasks, leveraging web search, data collection, and structured report generation.
Translates English or Japanese tech articles into natural, fluent Chinese using a three-step process: direct translation, issue identification, and reinterpretation. Preserves Markdown formatting and keeps technical terms and brand names in their original form.
A structured multi-agent workflow for thorough GitHub PR code reviews using 5 parallel specialized reviewers, confidence scoring, and false positive filtering to produce high-signal, actionable feedback.
A meta-skill for creating, testing, and iteratively improving AI agent skills through structured workflows including draft writing, evaluation benchmarking with quantitative assertions, user feedback loops, and description optimization for better trigger accuracy.