claude-skills and .claude
These are ecosystem siblings—both are configuration/plugin systems that extend Claude's capabilities through skills and agents, with A offering a curated marketplace-distributed collection while B provides a local ~/.claude config framework for organizing and auto-activating resources from multiple sources.
About claude-skills
alirezarezvani/claude-skills
+180 production-ready skills & plugins for Claude Code, OpenAI Codex, and OpenClaw — engineering, marketing, product, compliance, C-level advisory, and more. Install via /plugin marketplace.
Skills and agents are packaged as modular instruction sets with structured SKILL.md files, 268 stdlib-only Python CLI tools, and conversion scripts that auto-translate to native formats (`.mdc` for Cursor, `CONVENTIONS.md` for Aider, `.kilocode/rules/` for Kilo Code, etc.). The architecture separates skills (task execution), agents (task orchestration), and personas (cross-domain reasoning), all installable via plugin marketplaces or `./scripts/convert.sh --tool all` for multi-tool deployment. Designed to work across Claude Code, OpenAI Codex, Gemini CLI, Cursor, Aider, Windsurf, and seven other AI coding agents without platform-specific refactoring.
About .claude
travisjneuman/.claude
The Ultimate Claude Code Toolkit — 119 skills, 67 agents, 101 marketplace repos (6,600+ community skills), 30 commands, 8 hooks. Drop-in ~/.claude config that auto-activates the right resources from your prompt. Zero config required.
Implements a dynamic router that automatically loads relevant skills, agents, and rules based on prompt context, with 12 MCP server configurations and a flagship GSD project management framework (v1.29) providing 57 structured workflow commands. Integrates with Claude Code via stdio transport and leverages 110 community marketplace repositories to extend capability across 40+ domains—from blockchain and edge computing to compliance and XR development. Lifecycle hooks (pre-commit safety guards, session triggers) and stack-specific rule sets enforce quality guardrails without manual activation.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work