# Fuel Check Lean

Quick food analysis from a photo — no Garmin stats, no weather, no MCP calls. Just the food, the time, and your scores.

## Arguments

`<image_path>` — Absolute path to a food photo. If no argument, use the most recent image in `~/.claude/channels/telegram/inbox/` (sorted by filename timestamp).

Examples:
- `/fuel-check-lean` — analyze the last Telegram photo
- `/fuel-check-lean /path/to/photo.jpg` — analyze a specific image

## Workflow

### Step 1: Resolve Image

- If argument provided, use it as the image path
- If no argument, find the most recent `.jpg` or `.png` file in `~/.claude/channels/telegram/inbox/` by sorting filenames descending
- If no images found, tell the user to send a food photo via Telegram first

### Step 2: Read and Verify

- Read the image file
- If NOT food/drink: reply "That's not food." and stop
- If it IS food/drink: identify what it is (be specific — brand, flavor, size if visible)

### Step 3: Identify and Estimate

One line each:
- **Item:** what it is
- **Calories:** approximate
- **Key macros:** protein, carbs, fat, sugar, sodium — whatever matters most for this item

### Step 4: Score (6 Dimensions)

Present a compact table scoring each dimension out of 10 with a short justification. Use only time of day and the food itself for context — no Garmin data, no weather.

| Category | Score | Why |
|---|---|---|
| **Context Requirement** | X/10 | Based on time of day and what this food typically serves (energy, recovery, comfort, habit) |
| **Health** | X/10 | Nutritional quality |
| **Satiety** | X/10 | Will this fill you up? |
| **Taste** | X/10 | How good does this taste? Be honest |
| **Bloat** | X/10 | Higher = less bloat |
| **Addiction** | X/10 | How engineered is this to hook you? |

### Step 5: One-Line Read

One sentence. Opinionated. No hedging. That's it.