By hamsurang
Run a slash command to review any SKILL.md file against best practices, receiving a structured pass/fail report with warnings, errors, and prioritized recommendations before publishing or deploying to the marketplace.
npx claudepluginhub hamsurang/kit --plugin skill-reviewA community-driven plugin & skills marketplace for Claude Code
claude plugin marketplace add https://github.com/hamsurang/kit
# List all plugins from the marketplace
claude plugin list
# Search by keyword
claude plugin search <keyword>
claude plugin install <plugin-name>
Example:
claude plugin install vitest
# Update a specific plugin
claude plugin update <plugin-name>
# Refresh all plugins from the marketplace
claude plugin marketplace update hamsurang/kit
Or install individual skills without adding the marketplace:
npx skills add hamsurang/kit
Scaffold a new plugin interactively:
git clone https://github.com/hamsurang/kit
cd kit
bash scripts/scaffold-plugin.sh
→ Contributing guide · Plugin spec
| Plugin | Description | Author |
|---|---|---|
| vitest | Auto-invoked skill for writing, debugging, and configuring Vitest tests in Vite-based projects | minsoo.web |
| skill-review | Slash-command skill that reviews any SKILL.md against best practices and outputs a structured pass/fail report | minsoo.web |
| gh-cli | Auto-invoked skill for working with GitHub from the command line using the gh CLI tool | minsoo.web |
| personal-tutor | Adaptive technical tutoring skill that builds a persistent knowledge graph and learner profile across sessions | minsoo.web |
| deepwiki-cli | Query GitHub repository wikis via DeepWiki CLI without MCP token overhead. | minsoo.web |
| obsidian | Work with Obsidian vaults — search, create, edit, organize notes and manage frontmatter via obsidian-cli and Obsidian Headless | minsoo.web |
| library-analyzer | Analyze open-source libraries for contribution readiness with parallel agents | minsoo.web |
| obsidian-brain | Archive learnings from Claude Code sessions to Obsidian vault as Zettelkasten notes and use vault knowledge as conversational context | minsoo.web |
| velog-cli | Manage velog.io blog posts from the terminal — create, edit, publish, and browse posts via the velog CLI | minsoo.web |
| hamsurang-slide | Hamsurang brand HTML presentation generator. Soft Modern design with light/dark themes, 14 slide types, 4-color code highlighting. | Sonny |
Have a plugin to share? See Contributing.
The deepwiki-cli plugin uses a Rust CLI binary instead of the MCP protocol, delivering significant token and time savings.



Tested with 3 queries (React Reconciler, Next.js App Router, Tokio Async Runtime). CLI approach uses a single
deepwiki-cli askBash call; MCP approach performs the full 3-step protocol (initialize → tools/list → tools/call) via Streamable HTTP.
kit does not audit or sandbox plugins. Review any plugin — especially .mcp.json and shell commands — before installing. Only install from authors you trust.
MIT — see LICENSE
Tools for creating, auditing, and maintaining Claude Code skills. Includes /create-skill for scaffolding, /review-skill for quality checks, and /audit commands for bulk verification. Use when: building new skills, maintaining skill quality, or forking claude-skills repo.
Share bugs, ideas, or general feedback.
Documentation and authoring workflow router: audit docs vs code drift, sync docs after changes, optimize prompts and SKILL.md files, validate GLFM and Markdown formatting, summarize files/URLs/images with fidelity enforcement. Use when: docs are out of date, CLAUDE.md needs improving, SKILL.md needs optimizing, checking if documentation matches code, summarizing files or URLs.
Build, audit, and improve Claude Code skills with research-first architecture and opinionated structure
Evidence-based agent skills compiler with progressive capability tiers (Quick/Forge/Forge+/Deep).
Evaluate Agent Skill design quality against official specifications and best practices. Use when reviewing, auditing, or improving SKILL.md files and skill packages.
Produce and review technical documentation that is clear, actionable, and maintainable for both humans and agents, including contributor-governance files and agent instruction files.