I built a system with Claude Code that tracks your TikTok competitors, finds their most viral videos, and generates content ideas + scripts you can film right away. No more guessing what to post. No more staring at a blank screen. You add your competitors → the system scrapes their best content → AI analyzes the patterns → you get 5-10 video concepts tailored to your niche. Each one comes with a hook, structure, and CTA. It runs as a web app on your laptop. The whole thing was built with Claude Code. 👉 GitHub repo: https://github.com/melnikoff-oleg/tiktok-ai 👉 Full step-by-step install guide attached below — covers VS Code, Claude Code, API keys, and running the pipeline. If you get stuck, drop a comment. I'll personally help you get it running.

This guide covers the full setup from zero: installing VS Code, setting up Claude Code, connecting to the API, and the exact instructions I give Claude Code to scrape competitors and generate the report. No coding experience needed. If you can copy-paste, you can do this. What you'll build: - A system that scrapes every video from your competitors (titles, views, thumbnails, dates) - Automatic outlier detection that finds videos that got way more views than usual for that channel - Full transcript analysis of the top-performing videos - A beautiful HTML report with actionable insights, sorted by priority What you'll need: - A laptop (Mac or Windows) - ~$5/month for Apify (the scraping tool) - ~$20/month for Claude API credits (the AI that does the analysis) - 30-60 minutes of your time Let's go. STEP 1: Install VS Code VS Code is a free code editor. You won't be writing code, it's just where Claude Code lives. 1. Go to code.visualstudio.com 2. Download for your operating system (Mac / Windows / Linux) 3. Install it and open it 4. Open a terminal inside VS Code: press Ctrl+` (backtick) on Windows or Cmd+` on Mac. A terminal panel will appear at the bottom. That's it. You don't need to know how VS Code works beyond this. STEP 2: Install Claude Code Claude Code is Anthropic's CLI tool. It lets you have a conversation with Claude directly in your terminal, and Claude can read/write files, run scripts, and access APIs. On Mac: 1. Make sure you have Node.js installed. Open Terminal and run: node --version If you see a version number (like v20.x.x), you're good. If not, go to nodejs.org and install the LTS version. 2. Install Claude Code: npm install -g @anthropic-ai/claude-code 3. Verify it works: claude --version On Windows: 1. Install Node.js from nodejs.org (LTS version) 2. Open VS Code terminal and run: npm install -g @anthropic-ai/claude-code 3. Verify: claude --version STEP 3: Connect to the Anthropic API Claude Code needs an API key to talk to Claude.

I built a system that scans Twitter, news websites, and YouTube every morning — filters out the noise, clusters duplicate coverage into stories, and scores everything by relevance. No more doomscrolling 10 tabs to stay informed. In this video I show the full system live and walk you through building your own. Here's the quick setup guide: 1. Connect your data sources Claude Code supports external tools through MCP and API integrations. For this system you need: - Apify — scrapes Twitter/X accounts and YouTube channels. Get an API key at apify.com, add it to your .env as APIFY_API_KEY. Claude Code calls Apify actors via REST API to pull tweets and video metadata + transcripts. - Firecrawl — scrapes any website (blogs, news sites). Get an API key at firecrawl.dev, add it as FIRECRAWL_API_KEY. Claude Code uses it to grab article content from pages like techcrunch.com, openai.com/news, etc. - Kie AI — (optional) for generating visual assets from your trend data. Add as `KIE_AI_API_KEY`. 2. Define your sources Create a list of Twitter accounts and websites relevant to YOUR niche. Mine are 10 Twitter accounts (@OpenAI, @AnthropicAI, @karpathy, @sama, etc.) and 5 news sites. You can track any industry — crypto, biotech, marketing, whatever. 3. Write a Claude Code slash command Create a file at `.claude/commands/trends.md` that tells Claude to: - Fetch latest posts from your Twitter accounts (via Apify) - Scrape your news websites (via Firecrawl) - Filter to news only (drop opinions and commentary) - Cluster items about the same event into stories - Score each story 1-10 for relevance to your niche Then just run /trends in Claude Code and it does everything. 4. Generate a web interface Ask Claude Code to build you a simple Next.js page that displays your stories grouped by day, sorted by score. Mine has color-coded badges (green = high relevance, gray = low), expandable source lists, and voting buttons so the system learns what I care about over time.

American Bank. 35 million documents. No organization. 275 different document types mixed together. Manual classification would take years. AI did it in 2 weeks. THE DOCUMENT NIGHTMARE: Decades of files: - Loan applications - Bank statements - Tax returns - Asset verification - Compliance filings - Historical records All in shared drives. No naming convention. No searchable index. Compliance audits = nightmare. "Find all 2018 commercial loan docs" = weeks of manual searching. THE 4-NODE WORKFLOW: 1. Scan document folders 2. Convert PDFs and images to text 3. Classify by document type (loan app, tax return, etc.) 4. Extract key metadata (date, account number, dollar amount) 5. Build searchable database THE CHALLENGE: 35 million documents = expensive if you process everything. Solution: Process in batches, prioritize by business need: - Regulatory compliance docs first (immediate audit risk) - Active account documents second - Historical archives last Total processing: 2 weeks for critical docs, 3 months for full archive. THE IMPACT: Before: - Audit request: "All 2020 mortgage applications" - Time to comply: 2-3 weeks manual search - Cost: $18,000 in staff time After: - Same request: 15 minutes database query - Cost: $0 incremental - Audit compliance time: 94% reduction THE MORTGAGE PROCESSING USE CASE: Once classified, built automated loan processing: - Applicant uploads docs - System pulls income, employment, assets - Pre-fills underwriting system - Flags missing documents Loan processing: 30 days → 3 days. THE SALES ANGLE: Don't sell "document classification." Sell "you have 35 million documents you can't find when auditors ask. I make them searchable in 2 weeks." Compliance fear > efficiency desire. 📚 All templates library in Github What business has years of documents they can't search when they need them?

Hey guys, I've spent the last 7 days building this system. I helps you reverse engineer the strategy of any LinkedIn profile. So you can take a top creator from your niche, and have and get a super detailed report on how to replicate their success. Both their content strategy and sales strategy 💰 And I'm sharing this tool 100% free. I hope it helps on your personal branding journey. Use it here: https://www.buildauthority.ai/profile-analysis