claude-skills: 380 Skills for Claude Code, Codex & Cursor
Your AI agent just got a serious upgrade. The alirezarezvani/claude-skills repo ships 380+ skills, 30+ agents and 70+ custom commands for Claude Code, Codex, Gemini CLI, Cursor and 8 more coding agents. Engineering, marketing, product, compliance, C-level advisory, research, finance and daily productivity: all pre-built and ready to plug in.
⚡ What's inside:
• 380+ skills and 30+ agents across 12+ tools
• 70+ custom commands and scripts out of the box
• Customizable references and templates for real business tasks
• Works with Claude Code, Codex, Gemini CLI, Cursor and more
🛠 Quick start:
💡 Pro tip: load the marketing and finance skill packs and you get an AI analyst that writes briefs, models unit economics and preps C-level reports on autopilot.
🔗 Official repository
Your AI agent just got a serious upgrade. The alirezarezvani/claude-skills repo ships 380+ skills, 30+ agents and 70+ custom commands for Claude Code, Codex, Gemini CLI, Cursor and 8 more coding agents. Engineering, marketing, product, compliance, C-level advisory, research, finance and daily productivity: all pre-built and ready to plug in.
⚡ What's inside:
• 380+ skills and 30+ agents across 12+ tools
• 70+ custom commands and scripts out of the box
• Customizable references and templates for real business tasks
• Works with Claude Code, Codex, Gemini CLI, Cursor and more
🛠 Quick start:
git clone https://github.com/alirezarezvani/claude-skills.git💡 Pro tip: load the marketing and finance skill packs and you get an AI analyst that writes briefs, models unit economics and preps C-level reports on autopilot.
🔗 Official repository
PageIndex: RAG Without Vectors or Cloud Bills
Forget Pinecone, Weaviate and embedding invoices. PageIndex builds a document index through reasoning, not vectors. Result: sharper answers, less infra, zero vector DB costs.
⚡ Key features:
• Reasoning-based retrieval instead of cosine similarity
• Runs locally, no external vector DB needed
• Indexes document structure, not random chunks
• Perfect for PDFs, contracts, technical docs
🛠 Quick start:
💡 Pro tip: Drop a 200-page contract and ask questions by clause. PageIndex finds the right section by meaning, not keyword luck.
🔗 Official repository
Forget Pinecone, Weaviate and embedding invoices. PageIndex builds a document index through reasoning, not vectors. Result: sharper answers, less infra, zero vector DB costs.
⚡ Key features:
• Reasoning-based retrieval instead of cosine similarity
• Runs locally, no external vector DB needed
• Indexes document structure, not random chunks
• Perfect for PDFs, contracts, technical docs
🛠 Quick start:
git clone https://github.com/VectifyAI/PageIndex && pip install -r requirements.txt💡 Pro tip: Drop a 200-page contract and ask questions by clause. PageIndex finds the right section by meaning, not keyword luck.
🔗 Official repository
i-have-adhd: Stop Your AI Agent From Burying the Answer
Your coding agent dumps 40 lines of rambling and hides the actual answer at the bottom. This skill forces it to talk like someone with ADHD: short, direct, answer first.
⚡ What it does:
• Kills filler preambles and thinking-out-loud noise
• Answer always up top, details only on request
• Works as a skill for Claude Code, Cursor and other agents
• Saves tokens and your sanity
🛠 Quick start:
Drop the skill into your agent folder and restart the session.
💡 Pro tip: Wire it into your code review agent - no more scrolling walls of text to find one bug line.
🔗 Official repository
Your coding agent dumps 40 lines of rambling and hides the actual answer at the bottom. This skill forces it to talk like someone with ADHD: short, direct, answer first.
⚡ What it does:
• Kills filler preambles and thinking-out-loud noise
• Answer always up top, details only on request
• Works as a skill for Claude Code, Cursor and other agents
• Saves tokens and your sanity
🛠 Quick start:
git clone https://github.com/ayghri/i-have-adhdDrop the skill into your agent folder and restart the session.
💡 Pro tip: Wire it into your code review agent - no more scrolling walls of text to find one bug line.
🔗 Official repository
VibeVoice: Microsoft's Free Frontier Voice AI
Microsoft just open-sourced a frontier speech synthesis model. This is a direct replacement for pricey ElevenLabs and PlayHT, but running locally with zero character limits.
⚡ Key features:
• Realistic TTS on par with top paid services
• Runs 100% locally, no subscriptions or API keys
• Handles long texts and multi-speaker dialogues
• Open weights, fine-tune on your own voice
🛠 Quick start:
💡 Pro tip: Batch-voice your videos and podcasts without paying per character. Set it up once, generate unlimited audio forever.
🔗 Official repository
Microsoft just open-sourced a frontier speech synthesis model. This is a direct replacement for pricey ElevenLabs and PlayHT, but running locally with zero character limits.
⚡ Key features:
• Realistic TTS on par with top paid services
• Runs 100% locally, no subscriptions or API keys
• Handles long texts and multi-speaker dialogues
• Open weights, fine-tune on your own voice
🛠 Quick start:
git clone https://github.com/microsoft/VibeVoicepip install -r requirements.txt💡 Pro tip: Batch-voice your videos and podcasts without paying per character. Set it up once, generate unlimited audio forever.
🔗 Official repository