huawei-cloud-outfit-generate

Randomly generate one outfit-of-the-day (今日穿搭). A single call picks an overall style and a season, then draws exactly one item per category — 上装 (top) / 下装 (bottom) / 鞋履 (footwear) / 配饰 (accessory) — from a built-in wardrobe, so the look is a complete, coherent outfit. The style and season can be forced with --style / --season (they filter each category to matching items), output is readable Markdown by default with plain-text and JSON alternatives, and a --seed value makes any run reproducible. Implemented with the Python 3 standard library only; it does not call the Huawei Cloud CLI, any Huawei Cloud SDK, or any REST API, and it touches no cloud resource. Triggers include: "随机穿搭", "今日穿搭", "生成穿搭", "今天穿什么", "穿搭建议", "随机生成一套穿搭", "穿搭生成", "outfit generate", "random outfit", "outfit of the day", "daily outfit", "outfit generator", "generate an outfit".

Outfit Generator

Overview

This skill produces one random outfit-of-the-day per invocation. An overall style and a season are chosen first, then each category — 上装 / 下装 / 鞋履 / 配饰 — is filled with one item drawn from a built-in wardrobe (scripts/wardrobe.json). Every item is tagged with the styles and seasons it suits, so the result is a complete, readable look that actually respects the requested style / season.

• One call → one outfit — a single, self-contained 今日穿搭 covering all four categories.
• Conditional — --style and --season filter each category to matching items (relaxed gracefully, with a note, only when the wardrobe has no strict match).
• Reproducible — passing --seed makes a run deterministic; when omitted, a random seed is chosen and printed so the run can be reproduced later.
• Self-contained — pure Python 3 standard library, no network, no credentials.
• Demo data only — the wardrobe is bundled sample data; the output is a generated suggestion, not real merchandising data.

Prerequisites

• Python 3.8+ available as python3 (or python) on PATH.
• No Huawei Cloud credentials, CLI (KooCLI), SDK, or network access are needed; this skill is entirely local.
• scripts/wardrobe.json must be present (it ships with the skill) unless a custom wardrobe is supplied with --wardrobe.

Workflow

1. Load the wardrobe — load_wardrobe() reads scripts/wardrobe.json and returns the styles / seasons lists plus the per-category item pools (each item carries a name, its styles, and its seasons); --wardrobe may override it with a custom JSON file of the same shape.
2. Choose the look's frame — pick_outfit(seed, style, season, wardrobe) selects a style and a season with random.Random(seed) (or uses the values given on the command line, marking their *_source as cli).
3. Fill each category — for every category, items matching both the chosen style and season are preferred; the season constraint is relaxed first, then the style one, and any relaxation is recorded in fallbacks. One item is drawn from the resulting pool.
4. Render — render_markdown() / render_text() / render_json() turn the outfit into readable output (Markdown by default).
5. Emit — the result goes to standard output, or to a file when --output is given.

Core Commands

bash

# Default: one Markdown outfit, reproducible with the given seed
python3 scripts/generate_outfit.py --seed 20240501

# Plain text
python3 scripts/generate_outfit.py --seed 42 --format text

# JSON (machine-readable, for piping into other tools)
python3 scripts/generate_outfit.py --seed 42 --format json

# Force a style and a season (both accept aliases; see the usage guide)
python3 scripts/generate_outfit.py --seed 7 --style casual --season autumn

# Chinese labels work too
python3 scripts/generate_outfit.py --seed 7 --style 简约 --season 冬

# List the styles / seasons the wardrobe supports
python3 scripts/generate_outfit.py --list-styles
python3 scripts/generate_outfit.py --list-seasons

# Use a custom wardrobe and write the outfit to a file
python3 scripts/generate_outfit.py --wardrobe /path/to/wardrobe.json --output outfit.md

# Run the bundled unit tests
python3 -m unittest discover -s tests -v

The command-line entry point is scripts/generate_outfit.py; the built-in scripts/wardrobe.json is what the examples above read by default.

Parameter Confirmation

Parameter	Required	Description	Default
--seed	No	Integer seed for reproducible output	random (printed)
--style	No	Force the outfit style (casual/休闲, commute/通勤, sport/运动, vintage/复古, minimal/简约)	random
--season	No	Force the season to dress for (spring/春, summer/夏, autumn/秋, winter/冬)	random
--format	No	Output format: markdown, text, or json	markdown
--wardrobe	No	Path to a custom wardrobe JSON	scripts/wardrobe.json
--output	No	Write the outfit to a file instead of stdout	stdout
--list-styles	No	Print the available styles and exit	off
--list-seasons	No	Print the available seasons and exit	off

Reference Documents

Document	Description
references/usage-guide.md	Command recipes, reproducibility rules, styles/seasons, extending the wardrobe
references/verification-method.md	How the skill is verified (offline unit tests + spot checks)
references/acceptance-criteria.md	Acceptance criteria and out-of-scope items
references/dataflow-diagram.md	Mermaid data flow diagram
references/iam-policies.md	Permission notes (none required)

Notes

• The generator is read-only with respect to the cloud: it reads only its local wardrobe and writes only to standard output or a user-chosen --output file.
• The outfit is sample/demo content: a random styling suggestion, not real merchandising or sizing data.
• Output ordering is fixed: 上装 → 下装 → 鞋履 → 配饰.
• To expand the wardrobe, add {"name", "styles", "seasons"} entries to a category array in scripts/wardrobe.json; keep at least a couple of items per style and season so a strict (unrelaxed) match stays possible. No code change is required.