Work Smarter, Not Harder

Tools that help Content Marketers, Web Workers and SEO Practitioners get stuff done, better and faster with AI

Right, here's the thing. AI tools are powerful but they're also overwhelming. Documentation written by robots for robots. Enterprise solutions when you just need to get Tuesday's work done.

We build the bridges. MCP servers that connect your tools. Research-backed analysis that improves your content. Practical stuff that works.

No hype. No £997 courses. Just tools that help you work smarter.

Advanced content gap analysis using Query Decomposition and Keyword Fan-Out (Google's patented methodology). Tells you exactly what user queries your content covers - and what it misses. Built on academic research because I needed to understand how AI search engines actually evaluate content.

Works with Self-RAG validation, so no hallucinations. Every coverage claim includes exact quotes from your content.

npx @houtini/fanout-mcp@latest

SEO is dead. Long live GEO (Generative Engine Optimisation). This tool shows you exactly how AI search engines see your content - claim density, writing quality, E-E-A-T signals, extractability. Research-backed metrics that correlate with 40% higher AI citation rates.

Built it because traditional SEO metrics stopped predicting what gets cited by ChatGPT, Perplexity, and Gemini. Turns out AI search engines care about different signals.

npx @houtini/geo-analyzer@latest

Connect Claude to Google's Gemini with search grounding, deep research, and video analysis. Why? Because sometimes you need a second opinion. Or real-time information Claude doesn't have. Or you want to analyse YouTube videos for accessibility descriptions.

Works with Gemini 2.5 Flash - the fast, capable model that won't burn your API budget.

npx @houtini/gemini-mcp@latest

Real-time stock data, financial statements, analyst ratings, and market intelligence through Claude. Built because I needed market research capabilities without switching contexts. Get quotes, balance sheets, income statements, sector performance - all from conversation.

Free tier available. Works with Financial Modeling Prep's comprehensive API.

npx @houtini/fmp-mcp@latest

MCP server wrapping autonomous agents using the Anthropic Agent SDK. Provides capability abstraction and constants-driven prompt assembly for multi-step workflows. Still experimental, but it's solving the "how do I build complex workflows without spaghetti code" problem.

The architecture: clean separation between infrastructure (MCP) and agent logic (SDK). No hardcoded prompts.

The streamlined version of the original toolkit. Same power, less setup. Perfect when you need local AI analysis without the full feature set. One command, you're running.

Built for people who want the benefits of local LLM integration without maintaining fifty functions.

npm install -g @houtini/lite

Pick your flavour:

# Fan-Out content analysis (keyword variants + gap analysis)
npx @houtini/fanout-mcp@latest

# GEO analysis (AI search optimisation)
npx @houtini/geo-analyzer@latest

# Gemini integration (search grounding + research)
npx @houtini/gemini-mcp@latest

# Financial data (stocks, markets, analysis)
npx @houtini/fmp-mcp@latest

# Lightweight local AI toolkit
npm install -g @houtini/lite

Then configure in Claude Desktop:

{
  "mcpServers": {
    "fanout": {
      "command": "npx",
      "args": ["-y", "@houtini/fanout-mcp@latest"],
      "env": {
        "ANTHROPIC_API_KEY": "your-key-here"
      }
    }
  }
}

That's it. You're running.

• Content marketers trying to optimise for AI search engines
• SEO practitioners who know traditional metrics aren't enough anymore
• Web workers managing multiple tools and workflows
• Freelancers who can't justify another monthly subscription
• Anyone tired of context window limits and API rate limits

We're not building for enterprise. We're building for Tuesday morning when you need to ship something and the usual tools aren't cutting it.

I'm Richard. Twenty years on the web, two years deep in AI. Started building these tools because I needed them. Turns out other people need them too.

Every tool here solves a real problem I've hit. No theoretical "this might be useful" stuff. Just practical solutions to actual work problems.

Want the full story? Read the Houtini blog for guidance on getting started.

• 📖 Houtini Blog - Read guidance on getting started with MCP and AI tools
• 💬 GitHub Discussions - Share what you're building
• 🐛 Issues - Report problems, request features
• ✉️ Email - Direct line if you need it

Brought to you by Houtini.com - practical AI tools for Web Workers, Digital Marketers and SEO practitioners.

AI shouldn't be complicated. It shouldn't require a PhD or a massive budget. It should just help you work better.

Every tool we build follows these rules:

1. It solves a real problem
2. It works out of the box
3. It doesn't require a manual to understand

That's it. No manifestos. No disruption. Just tools that work.