My AI setup
The actual Claude Code configuration I use every day. Custom agents, MCP servers, skills, hooks, and the context layer that ties it all together.
Claude Code
Opus 4.6 · 61 components
Carousel Generator Agent
Generates LinkedIn carousel posts as self-contained HTML files. Exports to PNG slides and combined PDF. Dark and light themes with mobile optimized typography.
Claude Code Advisor Agent
A workflow optimization specialist for Claude Code itself. Audits your setup, suggests improvements, and helps you get the most out of CLAUDE.md files, skills, hooks, MCP servers, and context management.
Code Reviewer Agent
A senior code reviewer that analyzes your pull requests for bugs, security vulnerabilities, performance issues, and style consistency, then delivers structured, actionable feedback.
Copywriter Agent
A professional copywriter that combines classic direct response frameworks with modern digital strategies. Creates high converting content for LinkedIn, email, YouTube, ads, and more.
Crypto Analyst Agent
A deep crypto research agent covering on-chain analytics, tokenomics, narrative awareness, and macro context. Produces institutional quality analysis with a 5 part framework.
Design Critic Agent
A UI/UX specialist that reviews interfaces for visual quality, usability, accessibility, and psychological impact. Provides prioritized, actionable feedback grounded in design principles.
Email Marketing Strategist Agent
An email campaign specialist that creates welcome sequences, launch sequences, nurture campaigns, and re-engagement flows. Combines behavioral psychology with conversion optimization.
Greatest People Agent
Deep research agent for studying history's most successful people. Analyzes early career decisions, biggest levers, daily habits, mindset patterns, and synthesizes cross-person insights.
SDR Blueprint Agent
Complete SDR system architect. Conducts structured discovery, researches current tools and pricing, then generates a comprehensive HTML blueprint with the full outbound pipeline, tool stack, and implementation phases.
Senior Architect Agent
A software architect that reads before writing, maps before moving, and follows the codebase rather than personal preferences. Approaches every task with the discipline of maintaining large codebases for years.
Stock Analyst Agent
A deep equity research agent that pulls live market data, analyzes financials, compares valuations to peers, and produces comprehensive investment reports with bull, base, and bear case scenarios.
Successful Companies Agent
Deep research agent for studying the most successful companies in history. Analyzes origin stories, growth engines, competitive moats, founder decisions, and synthesizes cross-company patterns.
GitHub MCP Server
Connect your AI agent directly to GitHub. Manage repositories, pull requests, issues, code search, branches, and releases without leaving your terminal.
Notion MCP Server
Connect your AI agent to your Notion workspace. Search pages, create content, update databases, add comments, and organize your knowledge base from your terminal.
Slack MCP Server
Connect your AI agent to Slack. Read channels, search conversations, post messages, manage threads, and turn team discussions into structured action items.
Firecrawl MCP Server
Give your AI agent the ability to search the web, scrape any webpage, crawl entire sites, and extract structured data. Works on JavaScript heavy single page apps that simple scrapers cannot handle.
Brave Search MCP Server
Give your AI agent access to Brave's independent search engine. Web search, local business search, news, images, videos, and AI summaries with a generous free tier.
Google Workspace MCP Server
Connect your AI agent to Google Calendar. Check your schedule, find free time, create events, resolve conflicts, and plan your day without opening a browser.
Perplexity MCP Server
Give your AI agent access to Perplexity's search engine for real time information, sourced answers with citations, and deep multi-source research. Four tools ranging from quick lookups to thorough investigations.
Puppeteer MCP Server
Give your AI agent hands to use a browser. Navigate pages, click buttons, fill forms, take screenshots, run JavaScript, and test UI flows, all from your terminal.
Stripe MCP Server
Connect your AI agent to your payment infrastructure. Look up customers, check subscriptions, create invoices, manage products, issue refunds, and monitor revenue without opening the Stripe dashboard.
Todoist MCP Server
Connect your AI agent to your task management system. Create tasks from conversations, structure them by priority and deadline, plan your day, and track progress, all through natural language.
Yahoo Finance MCP Server
Give your AI agent access to real time stock prices, financial statements, earnings dates, dividends, analyst recommendations, and financial news. No API key required.
Obsidian MCP Server
Connect your AI agent to your Obsidian vault. Search notes, create and update content, navigate your knowledge graph, and bridge your local knowledge base with your AI workflow.
Claude in Chrome
Browser automation directly from Claude Code. Navigate pages, click elements, fill forms, take screenshots, and record GIFs in Chrome.
BrowserMCP
Browser automation via the BrowserMCP package. Navigate, click, type, screenshot, and interact with web pages programmatically.
Docker MCP Gateway
Manage Docker containers, images, and networks via the official Docker MCP gateway.
Hostinger
Manage VPS instances, domains, DNS records, and SSL certificates on Hostinger infrastructure.
Dokploy
Deploy and manage applications on VPS through Dokploy. Control deployments, environment variables, domains, and databases.
n8n
Workflow automation platform. Search, execute, and monitor n8n workflows directly from Claude Code.
Supabase
Backend database and auth for application development. Query tables, manage auth, and interact with Supabase projects.
Figma
Inspect Figma designs, extract design tokens, take screenshots of frames, and access design variables directly from your terminal.
CoinMarketCap
Real time crypto prices, market cap data, technical analysis, trending narratives, and upcoming macro events.
Transcript API
Extract transcripts from YouTube videos and podcasts. Convert audio content into searchable, quotable text.
Auto Lint Hook
Automatically runs your linter after Claude edits code files, catching style violations before they compound.
Auto Format Hook
Runs Prettier on saved files after every edit, keeping formatting consistent without manual intervention.
Pre-Commit Check Hook
Validates your code before Claude commits, running type checks, tests, or custom scripts to prevent broken commits.
CLAUDE.md Files
The project instruction file that Claude Code reads automatically. Define coding standards, architecture notes, common commands, and project context so Claude understands your codebase from the start.
Knowledge Base
A structured collection of markdown files that give Claude Code deep context about your domain, processes, and accumulated decisions. The difference between a generic assistant and one that understands your business.
Memory System
Claude Code's built in persistent memory that automatically stores user preferences, feedback, project context, and references across sessions. Learns from your corrections and remembers your patterns.
Git History Context
Claude Code reads your git history, blame, and diff information to understand how code evolved, who changed what, and why. The codebase's built in documentation layer.
Environment Variables
Configuration values that connect Claude Code to external services. API keys, database URLs, and service credentials that enable MCP servers and API integrations.
Codebase as Context
Your actual source code, file structure, and project configuration are the richest context source. Claude Code reads files, follows imports, and builds a mental model of your architecture on the fly.
settings.json
The user level settings file for Claude Code. Controls permissions, environment variables, hooks, and MCP server configurations that apply across all projects.
claude.json
The global configuration file for Claude Code sessions. Stores MCP server definitions, project specific overrides, and session metadata. The central registry for all external integrations.
Model Configuration
Control which Claude model powers your sessions and agents. Choose between Opus for maximum capability, Sonnet for speed, or Haiku for lightweight tasks.
Permission System
Fine grained control over which tools Claude Code can use without asking. Set global defaults, project specific overrides, and per tool rules to balance safety with workflow speed.
Project Structure
How to organize your Claude Code ecosystem files across global settings, project CLAUDE.md files, and shared resources. The directory layout that makes everything discoverable.
Agent Frontmatter
The YAML frontmatter format that defines custom agents. Controls the agent name, description, model, tool access, and display color. Must start on line 1 of the .md file.
Commit Skill
/commitA structured git commit workflow that reviews your changes, writes clear commit messages, stages appropriate files, and commits safely. Triggered with /commit.
Prompt Writer Skill
/prompt-writerA meta-skill that transforms rough ideas into polished, effective AI prompts for any tool: image generators, chatbots, coding assistants, video generators, and custom agents. Triggered with /prompt-writer.
Review PR Skill
/review-prA structured pull request review process that reads all changed files, analyzes for bugs, security, performance, and style, then delivers feedback with severity ratings. Triggered with /review-pr.
Project Decomposition Assistant
/pdaBreaks any project into phases, tasks (30-120 min), and MVAs (1-15 min). Applies Pareto 80/20 prioritization to focus on highest impact work first.
NotebookLM Skill
/notebooklmFull Google NotebookLM API integration. Create notebooks, add sources (URLs, YouTube, PDFs), generate AI podcasts, and download artifacts.
PRD Generator
/prdGenerate a Machine Optimized PRD designed for AI coding agents. Produces structured requirements that agents can execute directly.
Stock Analysis
/analyze-stockTriggers the stock-analyst agent for deep equity research. Reads the finance knowledge base, fetches live market data, and writes a comprehensive analysis file.
Crypto Analysis
/analyze-cryptoTriggers the crypto-analyst agent for deep token and protocol research. On-chain analytics, tokenomics, and macro framework analysis.
Morning Market Brief
/market-briefGenerate a concise morning market brief covering major indices, sector movements, notable earnings, and macro events using live web research.
Continue Prompt
/continueGenerate a continuation prompt file so work can be seamlessly resumed in a new Claude Code session. Captures task state, files, decisions, and next steps.
Todoist Integration
/todoistTodoist project management integration with pre-resolved project IDs, section IDs, labels, and MCP tool usage patterns for efficient task management.
Visualize Skill
/visualizeTransform any text source (book chapter, article, essay) into a premium visual HTML page with structured sections, typography, and optional personalization.