An AI-powered personal planning system using Cursor and markdown files
This is your intelligent personal planning system built on a simple idea: you describe your work in natural language, and the AI organizes everything for you.
- π
/daily-update- Share what you did today (AI learns and organizes everything) - π
/daily-briefing- Get your morning briefing (AI tells you what to focus on) - β
/create-focus-area- Create new tracking systems for anything you want
The system is flexible and extensible - these are starting points. You can add more commands or focus areas as needed.
Think of this like onboarding a new assistant - the AI needs context about you and your work!
On your very first /daily-update, don't hold back. Share:
About You & Your Role:
- What's your job title and responsibilities?
- What team are you on?
- Who do you work with regularly?
- Who's your manager?
Your Current Projects:
- What are you working on right now?
- What's the context and goals for each project?
- Any deadlines or milestones?
- Who are the stakeholders?
Your Goals:
- What are you trying to achieve? (Promotion, project completion, skills, etc.)
- What matters most to you professionally?
- Any upcoming performance reviews or important dates?
Your Work Patterns:
- When are you most productive?
- What kinds of meetings do you have?
- Common collaborators or stakeholders?
Example first update:
I'm a Senior Software Engineer on the Platform team, reporting to Jane Smith.
I'm currently leading the API microservices migration - we're breaking down
our monolith into independent services. Target is Q1 2025 completion.
Also working toward promotion to Staff Engineer - hoping for mid-year promo
cycle. I mentor two junior engineers (Sarah and Mike) and I'm trying to build
more cross-functional visibility.
Regular meetings: Daily standups (9:30am), 1:1 with Jane (Tuesdays 2pm),
architecture reviews (Thursdays). Most productive in mornings 9-12.
Current priorities: API migration is #1, promotion prep is #2, team mentoring
is ongoing. Key stakeholders: Product team (always emailing about timelines),
SRE team (for deployment), Engineering leadership.
Don't worry about being too detailed - this context helps the AI understand what matters and make better suggestions!
- Daily: Type
/daily-update(5-10 min) - just your day's highlights - Mornings: Type
/daily-briefing(5 min) - get your personalized briefing - As needed: Type
/create-focus-areato track new things
The system gets smarter with each update!
The system comes with a few starter commands in .cursor/commands/:
/daily-update - Log your day
- You: Share what you worked on (free-form, natural language)
- AI: Reads each focus area's instructions
- AI: Processes your update through each focus area
- Result: Everything organized automatically
/daily-briefing - Get your morning brief
- AI: Reads each focus area's instructions
- AI: Generates briefing for each focus area (may use Google Calendar/Gmail MCP)
- AI: Combines into comprehensive morning brief
- Result: You know exactly what to focus on today
/create-focus-area - Track something new
- AI: Guides you through creating a new focus area
- AI: Generates proper
instructions.mdand structure - Result: New tracking system ready immediately
You can add more commands! Edit .cursorrules or create new command files in .cursor/commands/.
The system has focus areas - self-contained modules that each track something different.
Each focus area:
- Has its own
instructions.mdthat tells the AI what to do - Organizes its data however makes sense
- Processes daily updates independently
- Contributes its own section to daily briefings
You can add any focus areas you want! Track whatever matters to you.
See /focus-areas/README.md for examples and how to create your own.
personal-planner/
β
βββ daily-logs/ # Your daily updates
β βββ YYYY-MM-DD.md
β
βββ focus-areas/ # Modular tracking systems
β βββ README.md # Guide to creating focus areas
β β
β βββ [your-focus-area-1]/ # Each focus area is independent
β β βββ instructions.md # Tells AI what to do
β β βββ [data-files] # Organized however you want
β β
β βββ [your-focus-area-2]/
β β βββ instructions.md
β β βββ [data-files]
β β
β βββ [add-more-as-needed]/ # Infinitely extensible!
β
βββ .cursorrules # System configuration
1. You Just Talk
- No form filling
- No categorizing
- Just describe your day naturally
- AI figures out the rest
2. AI Learns Over Time
- Week 1: Basic tracking
- Week 2-3: Patterns emerge
- Month 2+: Personalized insights
3. Modular & Extensible
- Each focus area is independent
- Add new focus areas anytime
- Remove what you don't need
- Customize instructions per area
4. Private & Yours
- All data in markdown files
- Your private repo
- No external databases
- Full control
You: /daily-update
AI: How was your day?
You: Spent most of the morning finishing up the user service extraction for
the API migration - that's finally done! Had our weekly 1:1 which was really
valuable, discussed promotion timeline. Helped Sarah debug a race condition
in the payment system. Got spammed by JIRA emails again about tickets I'm
not even on. Tomorrow need to start on the notification service.
AI: β
Processed your update through your focus areas:
[Each focus area processes according to its instructions.md]
Focus Area 1:
- [What it extracted/updated]
Focus Area 2:
- [What it extracted/updated]
Focus Area 3:
- [What it extracted/updated]
π Patterns learned:
- [Any new learnings across focus areas]
β
Daily log saved to /daily-logs/2025-01-15.md
Tomorrow's suggested focus: [Based on what focus areas learned]
You: /daily-briefing
AI:
# Daily Briefing - Tuesday, January 16, 2025
[Each focus area generates its briefing section based on its instructions.md]
## [Focus Area 1 Briefing]
[Generated content based on that focus area's data and instructions]
## [Focus Area 2 Briefing]
[Generated content - might use MCP integrations like Google Calendar]
## [Focus Area 3 Briefing]
[Generated content with suggestions and reminders]
[Continues for all focus areas...]
---
Have a productive day! π
The specific content depends entirely on what focus areas you've created and
what their instructions.md files tell the AI to do.
Want to track something else? Easy!
Using the helper command:
/create-focus-area
The AI will guide you through creating a properly structured focus area.
Or manually:
- Create
/focus-areas/[name]/ - Write
instructions.md(tells AI what to do) - Create data files (organize however you want)
- Done! AI automatically includes it
See /focus-areas/README.md for complete guide, templates, and examples.
Commands are just markdown files in .cursor/commands/. To add your own:
- Create
.cursor/commands/your-command.md - Write instructions for the AI to follow
- Update
.cursorrulesto reference it - Use
/your-commandin Cursor
The system is infinitely extensible!
Each focus area has:
instructions.md - Tells the AI:
- What to look for in daily updates
- Where/how to store information
- What to show in briefings
- How to learn and improve
Data files - Organized however makes sense:
- By time (sprint, quarter, year)
- By category
- By project
- Whatever works for that focus area!
When you run /daily-update or /daily-briefing:
- AI finds all focus areas
- Reads each
instructions.md - Follows those instructions exactly
- Each focus area processes independently
For full integration with Gmail, Calendar, Docs, Sheets, Slides, and Drive:
Install: Google Workspace MCP Server
# Install via uvx (easiest)
uvx workspace-mcpWhat it enables:
- Time-blocking: Real calendar data, optimized schedules around meetings
- Email intelligence: Process Gmail for priorities and action items
- Document access: Read meeting notes, project docs from Google Docs
- Data integration: Access spreadsheets, presentations, drive files
- Context awareness: Link daily updates to actual emails, meetings, documents
Setup in Cursor:
Add to your MCP settings (.cursor/mcp.json or similar):
{
"mcpServers": {
"google_workspace": {
"command": "uvx",
"args": ["workspace-mcp"],
"env": {
"GOOGLE_OAUTH_CLIENT_ID": "your-client-id",
"GOOGLE_OAUTH_CLIENT_SECRET": "your-secret"
}
}
}
}See the Google Workspace MCP documentation for:
- Creating OAuth credentials
- First-time authentication
- Available services and tools
- Advanced configuration
All focus areas work without MCP! You'll get:
- β Daily update tracking
- β Brag sheet building
- β General time block suggestions
- β No real calendar integration
- β No email processing
- β No document access
MCP makes the system significantly more powerful but isn't required.
- Be conversational: Talk like you're telling a friend
- Include context: Why things mattered
- Note patterns: "Another useless meeting", "Really productive morning"
- Mention people: Collaborations and help given/received
- Be honest: Challenges, blocks, frustrations
- Week 1: System is learning
- Week 2-3: Patterns start emerging
- Month 2+: System knows you well
- The more you use it, the smarter it gets
- Edit any focus area's
instructions.mdto change behavior - Add focus areas for anything you want to track
- Remove focus areas you don't need
- System adapts to YOU
- Day 1: Do comprehensive first
/daily-updatewith full context (see onboarding section above) - Review what AI extracted and where it stored things
- Day 2: Morning
/daily-briefing- see what AI learned - Day 2: Evening
/daily-update- shorter, just today's work - Week 1: Daily updates (build the habit and let AI learn)
- Week 2: Review focus areas - check what's being tracked
- Anytime: Add new focus areas with
/create-focus-area - (Optional) Set up Google Workspace MCP for calendar/email integration
Q: Do I need to use all focus areas?
A: No! Use what's helpful. Remove or ignore others.
Q: Can I edit the AI's work?
A: Yes! All files are yours to edit.
Q: What if AI gets something wrong?
A: Edit the file, and AI will learn from it.
Q: Can I track personal stuff too?
A: Absolutely! Add focus areas for anything.
This system is built on simple principles:
- Consistency > Perfection: Quick daily updates beat perfect weekly ones
- AI Augments, Not Replaces: You describe, AI organizes
- Memory is Powerful: Systems that learn get smarter
- Privacy Matters: Your data stays yours
- Modular is Better: Add what you need, ignore the rest
- Knowledge Share First: Treat the system like a new hire - onboard it properly!
Start with your first update:
/daily-update
First time? Give the AI context - share about your role, projects, goals, work patterns. Think of it like onboarding a new assistant who needs to understand your world.
After that? Just 5-10 minutes daily describing what you did. The AI learns and gets smarter over time.
The system is flexible - use what works, change what doesn't, add what's missing. It adapts to YOU.
Good luck! π―