mirror of
https://github.com/tiennm99/ai-coding-workflow-labs.git
synced 2026-09-19 02:22:32 +00:00
tiennm99
3336f04dff
Adds the full initial course content including: - START-HERE.md with navigation and teaching instructions - 12 lesson scripts covering delegation, frameworks, research, automation, and more - Scenario files for customer-feedback, done-framework, file-chaos, first-task, large-corpus, receipts, and strategy-notes exercises
153 lines
3.9 KiB
Markdown
153 lines
3.9 KiB
Markdown
# Lesson 5: The Inbox Pattern & Your First Skill
|
||
|
||
So far, we've done one-time organization.
|
||
|
||
But what about ongoing file management? Files come in constantly.
|
||
|
||
Let's build a system.
|
||
|
||
WAIT: Familiar with "inbox zero" for email?
|
||
|
||
USER: Yes / Sort of / No
|
||
|
||
---
|
||
|
||
Same idea for files:
|
||
|
||
- **inbox/** – Drop files here. Unprocessed. The chaos landing zone.
|
||
- **processed/** – Files I've handled, organized by type or date.
|
||
- **outputs/** – New things I create (reports, summaries, spreadsheets).
|
||
- **reference/** – Read-only context. Brand guidelines, templates, things I should read but not modify.
|
||
|
||
This is the folder structure I recommend for any Cowork project.
|
||
|
||
WAIT: Where would YOU drop a new receipt that just came in?
|
||
|
||
USER: [Says inbox]
|
||
|
||
Right. inbox/ is the landing zone. Everything starts there, then flows through the system.
|
||
|
||
---
|
||
|
||
Let me create this for you.
|
||
|
||
ACTION: Create the four folders (inbox/, processed/, outputs/, reference/) with README.md in each explaining purpose
|
||
|
||
ACTION: Create a root README.md explaining the system
|
||
|
||
Done. You now have a file processing system.
|
||
|
||
WAIT: Check your Artifacts – see the new structure?
|
||
|
||
USER: Yes
|
||
|
||
---
|
||
|
||
Now here's where it gets powerful.
|
||
|
||
A **Skill** is just a markdown file with instructions I follow.
|
||
|
||
You create it once. Use it forever.
|
||
|
||
Every time you want your inbox processed the same way, you tell me to use the skill.
|
||
|
||
This is how you get consistency without rewriting prompts.
|
||
|
||
WAIT: Want to create your first Skill?
|
||
|
||
USER: Yes
|
||
|
||
---
|
||
|
||
Let's write instructions for processing your inbox.
|
||
|
||
ACTION: Create skills/file-processor.md with:
|
||
|
||
```markdown
|
||
# File Processor Skill
|
||
|
||
## Purpose
|
||
Process all files in the inbox/ folder and organize them.
|
||
|
||
## Instructions
|
||
1. Read every file in inbox/
|
||
2. Determine what each file is by reading its contents
|
||
3. Move to processed/ in appropriate subfolder:
|
||
- processed/invoices/
|
||
- processed/receipts/
|
||
- processed/documents/
|
||
- processed/images/
|
||
- processed/other/
|
||
4. Rename files: YYYY-MM-DD-description.ext
|
||
5. Create outputs/processing-log-[date].md with all changes
|
||
|
||
## Constraints
|
||
- Never delete files
|
||
- Never modify files in reference/
|
||
- Ask before creating new category folders
|
||
|
||
## Naming Convention
|
||
YYYY-MM-DD-brief-description.ext
|
||
Example: 2026-01-15-amazon-receipt.pdf
|
||
```
|
||
|
||
There's your File Processor skill. It's now a permanent part of your system.
|
||
|
||
WAIT: Take a look at skills/file-processor.md. What would you change about this skill for your own work? Different categories? Different naming?
|
||
|
||
USER: [Suggests changes or says it works for them]
|
||
|
||
[If changes]: Good instinct – you can edit that file anytime to customize it.
|
||
|
||
[If works]: You can always tweak it later as you learn what you need.
|
||
|
||
---
|
||
|
||
Let's test it. I'll drop some sample files in your inbox.
|
||
|
||
ACTION: Create 5 sample files in inbox/ (varied types – a receipt image, a document, a screenshot, etc.)
|
||
|
||
Now tell me to process your inbox using the skill.
|
||
|
||
You can say something like: "Read the file-processor skill and process my inbox."
|
||
|
||
WAIT: Try it.
|
||
|
||
USER: [Instruction to use the skill]
|
||
|
||
---
|
||
|
||
ACTION: Read skills/file-processor.md and follow instructions to process inbox
|
||
|
||
Done. Check processed/ and outputs/.
|
||
|
||
WAIT: How'd that work?
|
||
|
||
USER: [Response]
|
||
|
||
---
|
||
|
||
That skill now exists forever in this folder.
|
||
|
||
Any time you want to process files the same way, just tell me to use it.
|
||
|
||
You can edit the skill file to change behavior. Add categories, change naming, add constraints.
|
||
|
||
This is how you build a personal assistant that knows YOUR preferences.
|
||
|
||
WAIT: Ready for research synthesis?
|
||
|
||
USER: Yes
|
||
|
||
---
|
||
|
||
**What you learned:** The Inbox Pattern for ongoing file management. Skills for reusable instructions.
|
||
|
||
**What you built:** A file processing system with your first Skill.
|
||
|
||
**Key insight:** Skills = instructions I follow. Create once, use forever.
|
||
|
||
**Next up:** Research synthesis – turning piles of documents into insights.
|
||
|
||
**To continue:** Start a fresh Cowork session and say: *"Read START-HERE.md and start lesson 6"*
|