I had the same problem. So I built a workflow.

One command. Five minutes. My entire day planned, calendar blocked, and yesterday's patterns analyzed.

What is this thing?

#

It's a multi-stage workflow built in Obsidian using Claude Code slash commands. Think of it as a personal assistant that actually knows your context.

Here's what it does automatically:

1. Reviews the last 3 days - Creates a vivid reconstruction of what actually happened (with exact timestamps)
2. Morning check-in - Asks reflection questions and creates a daily note
3. Reviews your goals - Generates contextual questions based on deadlines and progress
4. Schedules your day - Creates calendar events and a time-blocked task list
5. Logs everything - Tracks patterns so you see what's actually working

You run /morning-routine:main and it handles the rest.

No manual journaling. No staring at blank pages. No "what should I do today" paralysis.

Why I built this

#

Every morning I had to:

• Remember what I did yesterday
• Figure out what matters today
• Manually create calendar events
• Review goals scattered across different notes
• Actually motivate myself to do all this

That's too many steps.

So I automated it. Now the system remembers yesterday for me, asks smart questions about my goals, and creates calendar events automatically.

I just answer questions and make decisions. The workflow handles everything else.

How it actually works

#

Stage 1: Review yesterday (and the last 3 days)

#

Most people can't remember what they did yesterday The workflow reads your last 3 days of check-ins and creates a reconstruction with:

Specific timestamps: Not "worked in the morning" but "9:15 AM - started to log my expenses, got distracted by email at 9:47 AM"

Direct quotes: Your actual words from yesterday's reflections

Pattern analysis: "You mentioned being distracted 3 out of 3 days - always between 2-4 PM"

Action items you committed to: What you said you'd do yesterday vs what actually happened

Stage 2: Morning check-in

#

Once you know what happened yesterday, the workflow asks reflection questions:

• What's on your mind right now?
• What went well yesterday?
• What would make today great?
• Any blockers or concerns?

You answer however you want. Bullet points, full sentences, single words - doesn't matter.

The workflow accepts whatever you give it and creates a dated check-in file: 2026-01-07. Morning Check-in.md

Stage 3: Review goals

#

The workflow reads all your goal files and generates contextual questions based on:

Deadlines: "You have 4 days left on that blog post. Are you on track?"

Status: If a goal says "In progress" and you haven't logged anything in 5 days, it asks what's blocking you

Your own questions: If you ended a goal log with "Next review: What's the outline for section 2?" - it asks that

Patterns: If you keep mentioning the same blocker, it asks how to address it

You're not getting generic prompts like "What progress did you make?" You're getting questions that actually fit your context.

It uses extended thinking mode (AI reasoning) to analyze each goal and generate relevant questions.

After you answer, it updates your goal logs automatically with timestamps and your responses.

Stage 4: Schedule your day

#

Based on what you just said about your goals, the workflow creates:

Calendar events - 25-minute Pomodoro sessions scheduled sequentially using gcalcli (Google Calendar CLI)

A daily task file - Time-blocked list linking tasks to related goals

Realistic blocks -Actual time blocks accounting for breaks.

The calendar events appear immediately in your Google Calendar. You don't have to manually create them.

Your day is blocked before you start working.

Stage 5: Completion log

#

The workflow logs:

• Start time
• End time
• Total duration

Over weeks, you see patterns.

Using this for financial planning

#

You can use the goal tracking system to monitor financial habits and investment strategies:

The contextual questions adapt to your situation. If you set a goal to "Research REITs for passive income" and don't log progress for a week, it'll ask: "What's blocking the REIT research? Need better resources?"

I use it to track:

• My monthly stock pick research (3 new companies minimum)
• Weekly portfolio reviews (every Sunday)
• Daily market check-ins (15 minutes before work)
• Quarterly goal reviews (adjust FIRE number, check SWR assumptions)

The pattern analysis is incredibly useful. I noticed I skip my "review portfolio" goal every time the market drops more than 5%. That's emotional investing. The data made it obvious.

What you need to set this up

#

The core workflow works on any OS. Calendar integration works anywhere gcalcli runs (Linux/macOS/WSL2).

Installation overview

#

The complete workflow is available in my GitHub repository:

leviceroy/morning-routine-system

Shell

0

0

You can clone the entire vault structure or just grab the .claude/commands/morning-routine/ folder and adapt it to your needs.

Quick setup:

1. Clone the repository or download the vault files
2. Open it as a vault in Obsidian
3. Install Claude Code if you haven't already
4. (Optional) Set up gcalcli for Google Calendar integration
5. Run /morning-routine:main to start

The repository includes:

• All five workflow command files
• Template files for check-ins and goals
• Complete gcalcli setup instructions for Google Calendar integration
• Troubleshooting guide in .claude/README.md
• Sample goal files so you can see the format

If you want to customize it, the command files are just markdown with YAML frontmatter. Easy to modify.

Note: You'll need to set up gcalcli authentication once (it opens a browser for OAuth), then it works automatically from the command line.

What I learned using this daily

#

Is this overkill?

#

Maybe.

If you're happy with a paper journal or Apple Reminders and it works for you, stick with that.

But if you:

• Keep starting morning routines and quitting after a week
• Have goals scattered across different tools
• Struggle to remember what you did yesterday
• Spend 20 minutes every morning "figuring out" your day
• Want pattern analysis without manual tracking
• Need accountability for financial goals and investment research

Then yeah, this might be worth it.

The upfront setup takes an hour or two. But then you're running a personalized morning system that adapts to your life.

A workflow that knows your context and helps you act on it.

What's next?

#

I'm planning to add:

• Weekly review - Aggregate the 7-day patterns and show progress on goals
• Cross AI Platform - Make it work for AI platforms other than Claude Code
• Financial goal templates - Pre-built templates for FIRE goals, savings rates, portfolio reviews

Want to try it yourself? Download the complete workflow from the

leviceroy/morning-routine-system

Shell

0

0

Or just start simple: create a daily note every morning and write down what you did yesterday. You'd be surprised what patterns emerge.

What would you automate in your morning routine if you could? Possibilities are endless. Just think creatively

What's the difference?

The way they're asking.

Let me show you how this works, specifically for money stuff.

What actually is a prompt?

#

A prompt is basically your way of telling an AI model what you want. But there's a huge difference between asking like you're Googling something and actually giving the AI enough info to help you.

Bad version: "Help me budget"

Good version: "Act as a certified financial planner. I make $85,000/year, my fixed expenses are $3,500/month. Create a detailed budget using the 50/30/20 rule with specific dollar amounts and savings goals."

See what happened there? The second one gives the AI everything it needs. Who to be. What numbers to work with. How to format the answer.

Why this matters for your money

#

Financial decisions compound. You know this. A mediocre budget over ten years costs you thousands. A poorly thought-out investment strategy? Tens of thousands left on the table.

If you can get AI to give you actually useful financial advice - like, really tailored to your situation - you're basically getting a financial advisor for free. But only if you know how to ask.

The four things every good prompt needs

#

Alright, there's four pieces to this puzzle. Get these right and you'll get better advice than most people pay hundreds for.

1. Give it a persona

#

This is where you tell the AI what kind of expert you need it to be. And this totally changes the answer you get.

Here's what I mean:

Generic: "How should I invest?"

With a persona: "You're a fee-only fiduciary financial advisor who's helped middle-income people reach financial independence for 20 years. How should I invest?"

That second version? It's telling the AI to think like someone who has a legal duty to act in YOUR best interest. Someone who doesn't get paid commissions. Someone who knows the path to FI.

More examples you can use:

• "You're a tax specialist who helps high earners optimize their tax strategy..."
• "You're a retirement specialist who helps people in their 40s catch up on savings..."
• "You're a behavioral economist who gets why people struggle with spending..."
• "You're a debt elimination coach who's helped thousands pay off six-figure debts..."

Give your AI a personality. Seriously, it makes a massive difference.

2. Load it up with context

#

Context is the details about YOUR situation. This is what turns generic advice into something you can actually use.

No context: "Give me investment ideas under $10,000"

Loaded with context: "I'm 32, have $10,000 to invest. Already maxing out my tax-advantaged retirement accounts. Emergency fund's covered. I've got moderate risk tolerance, 30-year time horizon. Want to diversify outside retirement accounts. Give me five specific options with pros and cons."

The second one tells the AI:

• How old you are
• What you've already done
• How much risk you're okay with
• What timeframe you're working with
• Exactly what format you want

Throw in everything that's relevant:

• Your income and expenses
• Current debts and assets
• Time horizons (when do you need this money?)
• How much risk you can stomach
• Family situation
• Career stuff
• Your actual goals

The more you give it, the better it gets.

3. Tell it how to format the answer

#

This is where you get specific about how you want the information back. Format, length, tone. All of it.

Format example:

"Present the budget breakdown as:

1. A table with categories, amounts, and percentages
2. A ranked list of where I can cut spending
3. Action steps I can do this week"

Length example:

"Keep it under 300 words, but give me a detailed spreadsheet format for the calculations."

Tone example:

"Be professional but encouraging. I need straight talk, not sugar-coating. But also don't make me feel bad about my mistakes."

For financial stuff, you can ask for:

• Spreadsheet-ready formats
• Comparison tables
• Decision matrices
• Priority rankings
• Step-by-step plans

Whatever makes it easiest for YOU to use.

4. Show it examples (few-shot prompting)

#

This one's sneaky powerful. You literally show the AI exactly what you want by giving it an example.

Like this:

"I want you to analyze investment options. Format your response EXACTLY like this:

Investment Option: Global Stock Market Index Fund Risk Level: Moderate (6/10) Minimum Investment: $1,000 Expected Return: 7-10% annually Pros:

• Low expense ratio (typically 0.05-0.20%)
• Broad diversification across thousands of stocks
• Strong long-term historical performance Cons:
• No downside protection in bear markets
• Value fluctuates with market conditions Best For: Long-term investors (10+ years) comfortable with market swings

Now analyze these three investments using the same format: [your list]"

The AI will copy your structure exactly. Makes comparing options super easy.

Advanced stuff (Chain of Thought and Tree of Thoughts)

#

Okay so once you've got the basics down, there's some advanced techniques that'll blow your mind.

Chain of Thought (making it show its work)

#

This is where you tell the AI to think step-by-step before answering. It's like when your math teacher made you show your work, except here it actually helps.

Example for retirement planning:

"I'm 35 with $50,000 saved for retirement. Want to retire at 55 with $2 million. Before you give me a plan, think through step-by-step:

1. How much do I need to save monthly to hit $2 million by 55?
2. What investment return assumptions are we using? Are they realistic?
3. What could derail this plan? (Market crashes, job loss, inflation)
4. What adjustments could I make if I fall behind?

Show all your calculations and reasoning, then give me the recommendation."

Why this works:

• The AI catches its own mistakes
• You see the reasoning (builds trust)
• You actually learn something, not just get an answer

Tree of Thoughts (exploring multiple paths)

#

This one's wild. You have the AI explore multiple approaches, evaluate each one, then combine the best parts into one optimal strategy.

Here's an example for debt payoff:

"You're a financial counselor who specializes in debt elimination. I've got:

• $35,000 student loans at 5.5%
• $8,000 credit card debt at 18%
• $12,000 car loan at 4%
• $2,000/month surplus after minimums

Use Tree of Thoughts to find the best payoff strategy:

Step 1: Brainstorm three approaches:

• Branch A: "Debt Avalanche" - highest interest first (pure math)
• Branch B: "Debt Snowball" - smallest balance first (psychology)
• Branch C: "Hybrid" - balance transfer the credit card to 0%, then avalanche

Step 2: Evaluate each:

• Calculate total interest and timeline for each
• What are the psychological pros/cons?
• Risk factors (what if I lose my job?)
• Which builds momentum vs. saves most money?

Step 3: Combine the best parts into one "Golden Path" that balances:

• Math efficiency
• Psychological wins
• Risk management

Step 4: Give me the complete plan:

• Month-by-month payment schedule
• Total interest saved vs. just paying minimums
• Milestones to celebrate
• Backup plan if income changes

Show your reasoning for everything."

Why this is powerful: You get the benefits of multiple strategies combined. Pure optimization + psychological momentum + creative alternatives. All customized to YOUR situation.

Build a prompt library (I'm using Obsidian for that)

#

Here's what nobody tells you: once you find a prompt that works great, SAVE IT. Build a collection.

Good ones to save

#

Monthly budget review: "Act as my CFO. Review last month's spending [paste data]. Compare to my budget [paste budget]. Identify top 3 areas where I overspent, explain why, give me specific fixes. Format: Problem | Why It Happened | Solution | Expected Savings."

Investment rebalancing: "You're a fee-only advisor. Target allocation: 70% stocks, 25% bonds, 5% cash. Current allocation: [paste data]. Calculate exactly what I need to buy/sell to rebalance. Give me a specific action list with dollar amounts."

Tax optimization: "Act as a tax specialist for middle-income earners. Income: $X, current deductions: $Y, investments: $Z. Give me the top 5 ways to reduce my tax burden before year-end. Prioritize by savings amount and how easy they are to do."

Emergency fund check: "You're a risk management specialist. Monthly expenses: $X, current emergency savings: $Y, job stability: [level], dependents: [number]. Is my emergency fund adequate? If not, how much more and how fast should I build it?"

Where to keep them

#

Keep it simple:

• Text file (old school works)
• Note app (Obsidian, Notion, Evernote)
• Fabric tool (has an improve_prompt feature)
• Spreadsheet (organized by topic)

Tag them so you can find stuff:

• By topic (budgeting, investing, debt, taxes)
• By complexity (beginner, advanced)
• By time needed
• By output format

Making your prompts better over time

#

Your first attempts won't be perfect. That's normal. The key is getting better at it.

The cycle:

1. Try a prompt on a real question
2. Look at what you got back - was it useful?
3. Figure out what was missing
4. Adjust the prompt (more context, better formatting, whatever)
5. Try the new version

People like Daniel Miessler, Joseph Thacker, and Eric Pope have built entire frameworks around this. Miessler's Fabric tool has this improve_prompt feature that'll analyze your prompts and suggest how to make them better.

Common fixes:

• Too vague? Add more context and numbers
• Too long? Break it into multiple prompts
• Wrong tone? Adjust the persona and output requirements
• Missing stuff? Add examples
• Getting bad answers? Use Chain of Thought so you can see where it's going wrong

The real skill here is clarity

#

At the end of the day, this whole prompting thing is really about thinking clearly. When you get good at articulating exactly what you need from an AI, you also get better at:

• Asking good questions in real life
• Making clearer financial decisions
• Talking to actual financial advisors
• Understanding your own goals

Your action plan for 2026

#

Ready to actually use this stuff? Here's what to do:

Look, AI is changing how we deal with money. But it only helps if you know how to use it. Most people are getting garbage advice because they're asking garbage questions. And remember, it's not just for dealing with money, it's for everything else too.

You don't have to be most people.

Start building your prompt library today. Your financial freedom literally depends on asking the right questions the right way.

What's the first financial decision you're gonna tackle with this?

And if you're serious about financial freedom, you need to stop just tracking the past and start modeling the future.

That's where Python comes in.

Why your spreadsheet is holding you back

#

Financial independence means your passive income covers your living expenses. Simple concept but not that easy to execute.

You're planning for 30+ years of retirement. You've got inflation, market volatility and sequence of returns risk (the danger that the market crashes right when you retire). Unknown healthcare costs. Life not turning the way you expect.

A static spreadsheet with some formulas? That's not gonna cut it.

You need to model volatility and not assume stability. Worst case scenario!

Python lets you move from just reporting on your money to actually simulating and optimizing your future. It's the difference between looking in the rear mirror and having a GPS that shows you every possible route.

The four ways Python changes your financial game

#

Python and its libraries turn personal finance into a legit analytical project. Where YOU are the analyst.

1. Automated budgeting

#

First step to FI? Knowing exactly where your money goes.

Here's what Python can do:

My daily routine: I run Python scripts daily that track my spending and email me a report before I've had my morning coffee. I perfected that routine to my own liking.

No manual work. No forgetting to log stuff. Just automated tracking that runs in the background.

2. Portfolio optimization

#

This is where Python separates amateurs from pros.

Monte Carlo simulations

#

Instead of assuming "the market returns 7% every year" (which never happens), you can run 10,000+ simulations of different possible futures.

Each simulation represents a different way the market could play out. Some years are up 30%. Some are down 20%. Some are flat.

What you get:

• Actual probability your money lasts 30+ years
• Real success rates (like "95% chance of success")
• Quantified sequence of returns risk

This is what financial advisors charge thousands for. You can do it yourself.

Building better portfolios

#

Python lets you use Modern Portfolio Theory to find the optimal mix of assets for YOUR risk tolerance.

Pull historical data using free libraries like yfinance. Test different allocations. Customize everything based on when you want to retire and how much risk you can take.

Not some generic "60/40" portfolio everyone recommends. YOUR optimal mix.

3. Automated rebalancing

#

Your target is 80% stocks, 20% bonds. Market moves and now you're at 75/25.

Manually calculating what to buy or sell? A pain in the ass.

Python script? Tells you exactly how many shares to buy or sell to get back to your target. Accounts for trading costs. Optimizes which accounts to trade in for tax efficiency.

One click and you're done. Check out my portfolio rebalancing calculator.

4. Back-testing your plan

#

The biggest barrier to pulling the trigger on early retirement is Fear.

"What if I run out of money?" "What if the market crashes?" "What if I'm wrong about my safe withdrawal rate rule?"

Python lets you stress-test your exact plan against decades of real market history.

Want to know what would've happened if you retired in 2008? Run it.

Curious if 3.5% withdrawal is safer than 4%? Test both.

Wondering if a dynamic withdrawal strategy beats a static one? Back-test it.

This turns "hopeful guessing" into data-driven conviction.

You'll know if your plan would've survived the Great Depression, the dot-com crash, 2008, COVID. All of it.

How to actually start (my own framework)

#

You don't need to be a developer. You just need a plan.

Here's my PLOUTOS 4.0 framework:

Pro tip: Check out my calculators on LibreLeo for back-testing and simulations. I've already built a bunch of this stuff.

Why this actually matters

#

Financial independence is about maximizing freedom in your life.

Python is the tool that lets you:

1. Reduce anxiety by putting numbers on your risks instead of just worrying about them

2. Save time by automating tedious tasks that eat up hours every month

3. Gain confidence by stress-testing your plan against history's worst scenarios

You will have to put in the time to learn Python. There are some great learning platforms for python. For example Udemy.

But with AI tools now? It's easier than ever. AI tools can be a great coding companion.

The power you get over your financial future is worth the effort.

Your next steps

#

Stop reading. Start doing.

This week:

1. Install Python
2. Download your transaction history from your bank (CSV file)
3. Run a basic Pandas script to categorize your spending

This month: 4. Build a simple Monte Carlo simulation for retirement 5. Test your current portfolio allocation 6. Set up automated data pulls from your accounts

This quarter: 7. Create your first dashboard 8. Run back-tests on different withdrawal strategies 9. Optimize your portfolio based on actual data

The difference between people who talk about FI and people who achieve it? The ones who achieve it measure everything, test everything, and optimize relentlessly.

Python is how you do that without spending 40 hours a week on spreadsheets. Don’t get me wrong. Spreadsheets still have their place, but not as a standalone tool. They are far more powerful when combined with Python.

Start coding. Start automating. Accelerate your path to freedom.

Your future self will thank you.

Got questions about getting started with Python for finance? Drop them in the comments. I've been doing this for years and I'm happy to help.

My daemon is a queryable digital card that your AI can read using natural language. No forms, no manual searching—your AI talks to my daemon automatically.

My daemon API is intentionally public and accessible. I want your AI to query it. I want to be found by people who would benefit from connecting with me.

⚡ Quick Start (3 Steps)

#

Step 1: Add to Claude Desktop

#

Open your Claude Desktop config file:

• Mac: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

Add this server (no API key required):

{
  "mcpServers": {
    "chris-daemon": {
      "url": "https://mcp-daemon.chriswenk.workers.dev"
    }
  }
}

Restart Claude Desktop. You'll see a 🔌 icon indicating the server is connected.

Step 2: Add to Claude Code (CLI)

#

For Claude Code 2.0.69+, run this one command:

claude mcp add --transport http chris-daemon https://mcp-daemon.chriswenk.workers.dev

Then verify it's connected:

/mcp

Select chris-daemon and ensure it's enabled.

Step 3: Start Asking Questions

#

Open Claude Code and try:

Query chris-daemon to learn about Chris Wenk

What projects is Chris working on? Query chris-daemon

Query chris-daemon for book recommendations about financial independence

That's it. Your AI handles everything else.

💬 What You Can Ask

#

Ask your AI assistant in plain English:

✅ "What is Chris working on?"

• Queries my projects automatically

✅ "Find a good time to meet with Chris"

• Checks my location (Dubai) and daily routine

✅ "What books does Chris recommend?"

• Gets my reading list

✅ "Is Chris interested in Python automation?"

• Checks my mission, projects, and preferences for compatibility

✅ "Show me Chris's predictions about AI"

• Retrieves my future forecasts

No coding required. Your AI translates your questions into API calls automatically.

📡 All 13 Available MCP Tools

#

Your AI automatically picks the right tool based on your question.

🔧 For Developers: Direct API Access

#

Want to build your own tools? Use the API directly (no authentication required).

List Available Tools

#

curl -X POST https://mcp-daemon.chriswenk.workers.dev/ \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "method": "tools/list",
    "id": 1
  }'

Query Specific Tool

#

curl -X POST https://mcp-daemon.chriswenk.workers.dev/ \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "method": "tools/call",
    "params": {"name": "mission"},
    "id": 2
  }'

Python Example

#

import requests

def query_chris_daemon(tool_name: str) -> str:
    """Query Chris's daemon for information."""
    url = "https://mcp-daemon.chriswenk.workers.dev/"

    payload = {
        "jsonrpc": "2.0",
        "method": "tools/call",
        "params": {"name": tool_name},
        "id": 1
    }

    headers = {
        "Content-Type": "application/json"
    }

    response = requests.post(url, json=payload, headers=headers)
    return response.json()["result"]["content"][0]["text"]

# Usage
mission = query_chris_daemon("mission")
print(mission)

projects = query_chris_daemon("projects")
print(projects)

books = query_chris_daemon("favorite_books")
print(books)

🎯 Common Use Cases

#

1. Check Collaboration Fit

#

You: "I'm building a Python tool for portfolio tracking. Query chris-daemon to see if Chris would be interested."

AI: Queries mission, projects, preferences → Analyzes compatibility → Gives you a score and recommendation

2. Schedule a Meeting

#

You: "I'm in Dubai. Query chris-daemon and find a good time for a 30-minute call."

AI: Queries current_location + daily_routine → Calculates timezone difference → Suggests 3 optimal meeting times

3. Get Personalized Recommendations

#

You: "Query chris-daemon for book recommendations about financial independence."

AI: Queries favorite_books → Filters by topic → Returns relevant titles with explanations

4. Understand My Journey

#

You: "Query chris-daemon for Chris's narrative. I want to understand how he achieved financial independence."

AI: Queries narrative + about → Summarizes the journey from South Africa to Dubai to Financial Independence

🌐 Technical Details

#

Base URL: https://mcp-daemon.chriswenk.workers.dev Protocol: MCP (Model Context Protocol) 2024-11-05 - JSON-RPC 2.0 over HTTP Transport: HTTP (recommended for Claude Code 2.0.69+)

Authentication: None required (public access) Response Time: < 200ms average Availability: 99.9% uptime (Cloudflare Workers global network) Infrastructure: Two-tier architecture with service binding Rate Limit: Standard Cloudflare limits (designed for public use)

Architecture

#

Client (Your AI Assistant)
    ↓
Frontend MCP Server (Public - No Auth)
    ↓
    ↓ Service Binding (Internal)
    ↓ INTERNAL_AUTH_TOKEN
    ↓
Backend Worker (Private)
    ↓
daemon.md (Data Source)

Security:

• Frontend is completely public (no authentication required)
• Backend is completely private (only accessible via service binding)
• Internal authentication token protects service-to-service communication
• All data is intentionally public by design

❓ FAQ

#

Q: Do I need an API key? A: No! This server is completely public. No authentication required.

Q: Can I use this for my app? A: Yes! Build tools, integrations, discovery platforms—whatever helps people connect. It's free and open.

Q: What if I need private information? A: Contact me directly at whitealba@icloud.com

Q: Can I build my own daemon? A: Absolutely! The source code architecture uses Cloudflare Workers with MCP protocol. Contact me for implementation guidance.

Q: Does this work with Claude Code 2.0.69+? A: Yes! Just run: claude mcp add --transport http chris-daemon https://mcp-daemon.chriswenk.workers.dev

Q: Does this work with Claude Desktop? A: Yes! Add {"url": "https://mcp-daemon.chriswenk.workers.dev"} to your claude_desktop_config.json

🚀 Get Started Now

#

Option 1: Use Claude Code (Recommended)

claude mcp add --transport http chris-daemon https://mcp-daemon.chriswenk.workers.dev

Then in Claude Code:

1. Run /mcp to verify the server is connected
2. Ask: "What is Chris Wenk working on?"
3. Claude will automatically query the MCP server

Troubleshooting:

• If connection fails, try: claude mcp remove chris-daemon then re-add
• Run /doctor to check for configuration issues
• View detailed logs: claude --debug

Option 2: Use Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "chris-daemon": {
      "url": "https://mcp-daemon.chriswenk.workers.dev"
    }
  }
}

Restart Claude Desktop and look for the 🔌 icon.

Option 3: Build Your Own Tool

1. Use the curl or Python examples from the "For Developers" section above
2. Customize for your use case
3. Integrate into your workflow - it's completely free!

📬 Contact

#

Questions? Leave a comment and I'll reply API: https://mcp-daemon.chriswenk.workers.dev

The future of networking is AI-mediated discovery. My daemon is public and queryable. Use it. Build with it. Help enable the future.