Copy-paste prompts for the live build and Cowork workflow. Same text as the session deck. Homework at the bottom.
Claude Code prompts go in your terminal session pointed at Pennywise. Cowork prompts go in Cowork after the build.
Tap Copy prompt or select the text. Deck: Week 2 session slides.
You are helping me understand my own project before we change anything. Read the full Pennywise codebase and tell me, in plain English for a non-engineer: 1) what the app does end to end, 2) the main files and what each one is responsible for, 3) where and how expense data is stored, 4) anything that looks unfinished, fragile, or likely to cause problems later. Do not change any code. End with a short list of what you would need from me before adding a new feature.
I want to add three things to Pennywise: 1) import bank transactions from a CSV file the user uploads, 2) automatically categorise each transaction, 3) a monthly summary view showing total spend per category. Before writing any code, give me a step-by-step build plan. For each step tell me what you will change, which files it touches, and what I will be able to see or test when that step is done. Flag any decisions you need me to make. Keep the steps small enough that I can review each one. Do not write code until I reply 'go'.
Go. Build only step 1 from the plan. When you are done: 1) show me exactly what changed as a diff, 2) explain in one or two plain sentences what this step does, 3) tell me how to test it right now, 4) stop and wait for me before starting step 2. If you hit anything ambiguous, ask me instead of guessing.
Something is broken. Symptom: when I upload my CSV, the monthly summary shows nothing and I get an error on screen. Here is exactly what I see: [paste the error message and describe what you clicked]. Do not start rewriting yet. First find the most likely cause and explain it to me in plain English. Then propose the smallest fix that solves it. Once I say go, make only that fix, show me the diff, and tell me what the actual problem was so I learn from it.
Act as a product lead helping me document a feature I just shipped. The feature: Pennywise now lets a user upload a bank CSV, auto-categorises the transactions, and shows a monthly summary by category. Write a one-page PRD with these sections: Problem, User, What it does, Out of scope, Success (with one or two concrete signals). Keep it under one page, plain language, no jargon. Ask me any question you need before writing.
Research the expense-tracking space for me. Find three real tools that let users import transactions from a bank or CSV. For each one give me: the name, one line on who it is for, one thing it does better than my Pennywise feature, and one thing it does worse or is missing. Use current information. Then in two sentences tell me where a simple, private, self-built tracker like Pennywise actually has an edge.
Write a short LinkedIn post announcing that I just built CSV import and a monthly summary into my own expense app, Pennywise, during a bootcamp using Claude Code. Voice: direct, warm, first person, no hype, no em dashes. Structure: a one-line hook, two or three lines on what I built and why it was satisfying, one line on what I learned about directing an AI agent, and a soft invitation to follow along. Keep it under 120 words and give me two versions.
You have felt the basic loop. Here is the repeatable layer beneath it: memory, workflows, workers, and connections. Full syllabus lesson with setup steps, examples, and official doc links:
Open full guide on the syllabus →Install from code.claude.com, then cd into your local Pennywise or capstone folder and run claude. Claude Code works on files on disk. Export from Lovable first if needed.
curl -fsSL https://claude.ai/install.sh | bash, or VS Code extension).claude and log in..claude/.Markdown file loaded every session. Facts you would otherwise re-type: what the app is, build commands, never-break rules. Not long procedures (those go in Skills).
Run /init to draft from your repo. Edit, then commit CLAUDE.md to git.
npm run dev; amounts in cents; do not rewrite migrations.Folder .claude/skills/<name>/SKILL.md with YAML frontmatter (name, description) plus instructions. Claude auto-loads when the description fits, or you type /skill-name.
Unlike CLAUDE.md, skill content loads only when used, so long checklists are cheap until you need them.
--- name: new-feature description: Plan a feature step by step before writing code. --- 1. Read relevant code. Do not edit yet. 2. Write a step-by-step plan with files and tests. 3. Wait for "go". Build one step at a time. Show diff after each step.
Homework: create /new-feature and use it on your second feature.
Separate Claude instance with fresh context. Main chat stays focused; worker reports back.
Fast codebase search. Write/Edit denied. Skips CLAUDE.md to stay cheap. "Map how expense data flows" without cluttering your diff review.
Research for plan mode. Same read-only posture as Explore.
Your own workers: description, tool limits, optional preloaded skills. Safe default: read-only for research, main session makes edits after you approve.
Model Context Protocol links Claude Code to GitHub, Supabase, Slack, Notion, and more. Connect when you keep copy-pasting from another system.
For Pennywise: GitHub + database (read-only creds for exploration) are highest leverage. Add servers with claude mcp add or /mcp. Start with one or two, not everything.
PRDs, competitive research, LinkedIn copy. Same directing discipline: specific brief, check output, iterate. Attach notes and screenshots. Homework: publish a LinkedIn post, not just a draft.
| When | Reach for |
|---|---|
| Always true about the project | CLAUDE.md (+ /init) |
| A workflow you repeat | Skill (/new-feature) |
| Big read that would clutter chat | Subagent (Explore) |
| Data outside your files | MCP (GitHub, Supabase, …) |
| PRD, research, launch copy | Cowork |
| Share setup with your team | Plugin |
Session deck slides 23–29 cover this live. Official docs: Skills, Subagents, Memory, MCP.
Add one new feature to Pennywise using Claude Code, from plan to working, on your own. Then document it in Cowork.
Pick one (or propose your own):
Rule of thumb: pick something Lovable would struggle to do cleanly.
Ask Claude Code to write a short plain-English note in your project explaining how the feature works.
Time: 60 to 90 minutes. Stuck more than 15 minutes is a WhatsApp question, not a solo struggle.
Create a skill called new-feature that always plans before writing code, builds one step at a time, and shows me the diff before each step. Save it as a SKILL.md in this project.
Write a short LinkedIn post announcing that I just built [your feature] into my own expense app, Pennywise, during a bootcamp using Claude Code, and that I also saved my build method as a reusable Claude Code skill. Voice: direct, warm, first person, no hype, no em dashes. Structure: a one-line hook, two or three lines on what I built and why it was satisfying, one line on what I learned about directing an AI agent, and a soft invitation to follow along. Under 120 words, give me two versions.