← Back to Skills Marketplace
ivangdavila

Fashion

by Iván · GitHub ↗ · v1.0.0
linuxdarwinwin32 ✓ Security Clean
830
Downloads
5
Stars
2
Active Installs
1
Versions
Install in OpenClaw
/install fashion
Description
Style outfits, decode dress codes, and shop smart with body-aware advice, trend intelligence, and practical constraints.
README (SKILL.md)

When to Use

User needs outfit advice, wardrobe strategy, shopping guidance, or style problem-solving. Agent handles everything from daily dressing to event preparation, adapting to body type, budget, climate, and lifestyle constraints.

Quick Reference

Topic File
User-specific guidance users.md
Styling fundamentals styling.md
Fabric knowledge fabrics.md
Shopping intelligence shopping.md
Occasion dressing occasions.md

Core Rules

1. Ask Context Before Advising

Never give generic advice. First establish:

  • Body: Height, build, proportions (short torso? broad shoulders?)
  • Budget: Actual spending limit (€50/month ≠ €500/month)
  • Lifestyle: Job type, commute, physical activity, kids?
  • Environment: Industry dress codes, climate, cultural context
  • Constraints: Mobility needs, nursing, medical devices, sensory sensitivities

2. Work With What They Have

Before suggesting purchases, ask what's already in their wardrobe. Build outfits from existing pieces first. "Buy a blazer" is useless when they need to get dressed NOW.

3. Body Proportions Over Generic Types

Skip "apple/pear" labels. Ask specifics:

  • Shoulder-to-hip ratio
  • Torso vs leg length
  • Where waist naturally sits

Apply visual balancing: high-rise elongates short legs, V-necks balance broad shoulders. See styling.md for proportion rules.

4. Practical Constraints Are Non-Negotiable

Always factor in:

  • Mobility: Can they run, squat, sit for hours?
  • Care: Machine-washable or dry-clean? Ironing capacity?
  • Climate: Actual temperature + indoor/outdoor transitions
  • Budget: Not just price, but cost-per-wear math

5. One Recommendation, Not Options (When Asked)

If someone needs to get dressed in 5 minutes, give ONE answer, not five choices. Decision fatigue is real. Save options for when they're exploring.

6. Confidence Over "Flattering"

Never frame advice as "hiding" or "minimizing" body parts. Focus on what makes them feel powerful, comfortable, and like themselves. "This celebrates your shape" not "This hides your stomach."

7. Context-Specific Dress Codes

"Business casual" varies wildly:

  • Tech startup BC ≠ Law firm BC ≠ Finance BC
  • NYC smart casual ≠ Austin smart casual
  • Always ask industry, company culture, and specific venue

Adaptation Rules

For Different Bodies

  • Plus-size: See users.md — know actual brand size ranges, avoid "hide your body" defaults
  • Petite: Translate standard lengths (their "midi" = your maxi), prioritize proportion
  • Tall: Inseam/sleeve length sourcing, proportion balancing
  • Adaptive needs: Seated proportions, closure types, medical device access

For Different Lifestyles

  • Parents: Stain-camo fabrics, movement-friendly, 2-minute outfit decisions
  • Travelers: Wrinkle-resistant fabrics, layering systems, carry-on constraints
  • Budget-limited: Thrift strategy, repair skills, cost-per-wear prioritization
  • Minimalists: Uniform approach, single-answer decisions, no browsing

For Different Cultures

  • Non-Western: Traditional-modern fusion, modesty as mainstream, regional brands
  • Religious requirements: Build WITH the requirement, don't suggest removing it
  • Regional codes: What's appropriate varies by city, industry, generation

Shopping Traps

  • Cross-brand sizing varies wildly — Zara runs 1-2 sizes smaller than H&M
  • "Original price" is often fake — item may have never sold at that price
  • "Free returns" may mean fee deducted from refund
  • "Sustainable" without certifications (GOTS, B Corp) = probably greenwashing
  • Sale timing: winter coats cheapest in Feb, swimwear in Sept

Fabric Traps

  • Under 150 GSM t-shirts are see-through; 180+ is substantial
  • Cotton base layers in cold = dangerous (retains moisture)
  • "Vegan leather" = usually plastic with worse longevity
  • Linen wrinkles badly; merino wool travels well
  • "Dry clean only" is a lifestyle cost, not just a care label

Trend Guidance

Never recommend trends without lifecycle context:

  • Emerging: Runway only, not yet retail
  • Ascending: Street style adoption, entering stores
  • Peak: Fast fashion saturation — already over for early adopters
  • Declining: Ironic use only

State which phase. See styling.md for aesthetic distinctions (old money ≠ quiet luxury ≠ mob wife).

Usage Guidance
This skill is internally coherent and low-risk from a technical standpoint: it has no installers, no code to execute, and it doesn't ask for credentials. However, it is designed to collect personal body and wardrobe details (measurements, proportions, photos, medical/adaptive needs) to generate recommendations. Before using it, consider: 1) Are you comfortable sharing precise measurements or photos with the agent/model? If not, provide ranges or descriptions instead of exact numbers. 2) Avoid uploading identity-bearing photos or files unless you trust the platform's privacy policy — the skill itself does not declare storage/endpoint behavior, so data will be handled by the hosting agent/model. 3) Prefer textual descriptions (e.g., '5'4", long torso, broad shoulders') over full-name or photo if privacy is a concern. 4) Note the skill source is unknown; while instruction-only skills are low technical risk, check the platform's policy about where user data is sent and retained. If you need stronger guarantees, ask the provider about data retention or run similar advice locally without sharing sensitive details.
Capability Analysis
Type: OpenClaw Skill Name: fashion Version: 1.0.0 The OpenClaw AgentSkills bundle is benign. It consists entirely of markdown files providing fashion advice and instructions for the AI agent on how to deliver that advice. The `SKILL.md` instructions guide the agent on user interaction, ethical considerations, and knowledge application, without any directives for malicious actions like data exfiltration, unauthorized execution, or system access. The `_meta.json` explicitly states `"bins":[]`, indicating no external binaries are required, which limits potential shell injection risks. All content is aligned with the stated purpose of a fashion advice skill.
Capability Assessment
Purpose & Capability
Name, description, and included guidance files (users.md, styling.md, etc.) are consistent with a styling/shop-intelligence skill. It requests no binaries, env vars, or access outside of the agent context — nothing appears extraneous for the stated purpose.
Instruction Scope
SKILL.md instructs the agent to solicit detailed, often sensitive user context (height, proportions, wardrobe inventory, nursing/medical constraints). That is appropriate for tailored outfit advice, but it means the agent will ask for and process personal data. There are no instructions to read local files, environment variables, or to contact external endpoints beyond normal model usage.
Install Mechanism
This is an instruction-only skill with no install spec and no code files to execute. That minimizes supply-chain risk — nothing is downloaded or written to disk as part of installation.
Credentials
The skill declares no required environment variables, credentials, or config paths. The lack of requested secrets is proportionate to its functionality.
Persistence & Privilege
always:false (not forced into every agent), and disable-model-invocation:false is the normal default allowing autonomous invocation. The skill does not request system-wide changes or cross-skill credentials.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install fashion
  3. After installation, invoke the skill by name or use /fashion
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release
Metadata
Slug fashion
Version 1.0.0
License
All-time Installs 3
Active Installs 2
Total Versions 1
Frequently Asked Questions

What is Fashion?

Style outfits, decode dress codes, and shop smart with body-aware advice, trend intelligence, and practical constraints. It is an AI Agent Skill for Claude Code / OpenClaw, with 830 downloads so far.

How do I install Fashion?

Run "/install fashion" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Fashion free?

Yes, Fashion is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Fashion support?

Fashion is cross-platform and runs anywhere OpenClaw / Claude Code is available (linux, darwin, win32).

Who created Fashion?

It is built and maintained by Iván (@ivangdavila); the current version is v1.0.0.

💬 Comments