
Today vs proposal: table plus an editable chart
Bakery Central Kitchen Board Deck
An eight-slide board deck for Hearth & Grain, an illustrative five-shop bakery chain deciding whether to open a sixth shop with a central kitchen. Flour-white background, deep brown text and one wheat-gold accent, as the brief asked. Every title, table, diagram and both charts are native PowerPoint objects you can edit; the chart data opens with Edit Data, and the speaker notes show where each figure comes from.
How it was made
Recorded: a creator-reviewed session, November 12, 2026.
Models
- claude-opus-5-5 · anthropic · execution · Claude Code
Agent
Claude Code
Tools and APIs
- python-pptx
Build every slide, table and chart as native, editable PowerPoint objects.
- venv
Create an isolated Python environment to install python-pptx.
- Write
Write the deck build script and the deck notes.
- Read
Check the rendered slide previews between fixes.
- Monitor
Watch the remote preview render until it finished.
- Possible job runner
Receive the brief, render the slide previews with LibreOffice on a remote machine and hand back the result.
Ordered steps
Give Deck Maker a three-part brief
Fill in the agent's brief: who the deck is for and what they should decide, the facts and numbers to use, and the length and look. The agent built, previewed and checked the deck without follow-ups.
Who is it for, and what should they decide? The owners' board of Hearth & Grain, a five-shop neighbourhood bakery chain, deciding whether to open a sixth shop with a central kitchen that supplies all six. The facts and numbers to use (an illustrative scenario, not a real company): - Today: 5 shops, revenue SGD 4.2M/yr, 11% operating margin, each shop bakes on site. - Proposal: a central kitchen (SGD 650k fit-out) plus a sixth shop (SGD 280k), opening in 9 months. - Central baking cuts ingredient waste from 9% to 5% and frees 1.5 baker roles per shop. - Expected: revenue SGD 5.3M/yr in year 2, margin 14%; payback 26 months. - Main risks: delivery logistics before 6am, loss of "baked here" appeal, a 3-month dip during changeover. - Recommendation: approve, with a 2-shop pilot of central supply first. Length and look: eight slides, warm and simple: flour-white background, deep brown text, one wheat-gold accent. Editable charts.
Made with the Deck Maker skill
See the skillUS$19 for this skill.
Buy the Deck Maker skillOne payment, then you download it. It runs on your own machine, in Claude Code or Codex, with your own subscription.
By buying, you agree to our Terms and Refund Policy. Your purchase is sold through Link, a Stripe company, which takes your payment.
About
An eight-slide board deck for Hearth & Grain, an illustrative five-shop bakery chain deciding whether to open a sixth shop with a central kitchen. Flour-white background, deep brown text and one wheat-gold accent, as the brief asked. Every title, table, diagram and both charts are native PowerPoint objects you can edit; the chart data opens with Edit Data, and the speaker notes show where each figure comes from.
How it was made
An AI agent made this. Deck Maker, a Possible skill run in Claude Code with Claude Opus 5.5, got the brief shown here and nothing else: a fresh, empty workspace and no follow-ups. It built the deck with python-pptx, rendered slide previews with LibreOffice on a remote Linux machine, and fixed what it saw before handing it back. Possible recorded the session with possible capture. Nobody edited the deck by hand.
What to know
Deck Maker uses only the facts it is given. The brief states a 26-month payback, but its own revenue and margin figures give about 40 months; the deck shows both, marks the basis of the 26 months as TBD and does not pick one. Other missing inputs (ingredient spend, cost per baker role, pilot criteria, risk thresholds) are marked TBD rather than invented. Hearth & Grain is not a real company. The brief is public; the agent's standing instructions are not. Results vary between runs.
Exact prompt
Hidden characters, if present, appear as Unicode escapesWho is it for, and what should they decide? The owners' board of Hearth & Grain, a five-shop neighbourhood bakery chain, deciding whether to open a sixth shop with a central kitchen that supplies all six.
The facts and numbers to use (an illustrative scenario, not a real company):
- Today: 5 shops, revenue SGD 4.2M/yr, 11% operating margin, each shop bakes on site.
- Proposal: a central kitchen (SGD 650k fit-out) plus a sixth shop (SGD 280k), opening in 9 months.
- Central baking cuts ingredient waste from 9% to 5% and frees 1.5 baker roles per shop.
- Expected: revenue SGD 5.3M/yr in year 2, margin 14%; payback 26 months.
- Main risks: delivery logistics before 6am, loss of "baked here" appeal, a 3-month dip during changeover.
- Recommendation: approve, with a 2-shop pilot of central supply first.
Length and look: eight slides, warm and simple: flour-white background, deep brown text, one wheat-gold accent. Editable charts.What you need
- Claude Code, or another coding agent that can write files and run Python
- python-pptx
- PowerPoint, Keynote or LibreOffice to open and edit the .pptx
Download the result
- pptxThe editable 8-slide deck (PowerPoint, with speaker notes)↓
- markdownDeck notes: slide plan, derived arithmetic, assumptions and TBDs↓
- pngSlide 1: cover↓
- pngSlide 2: decision requested↓
- pngSlide 3: today vs proposal↓
- pngSlide 4: operating model↓
- pngSlide 5: investment and payback↓
- pngSlide 6: plan↓
- pngSlide 7: risks↓
- pngSlide 8: decision↓
Technical details and provenance
- Provider
- anthropic
- Model
- claude-opus-5-5
- Agent
- Claude Code
- Published
- Publication
- Made by Possible
- Source
- Original source ↗