← Back to Skills Marketplace
xavierjiezou

virtual-tryon-scorer

by TuringParty · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
113
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install virtual-tryon-scorer
Description
Score and evaluate virtual try-on (VTON) results by analyzing identity preservation, garment fidelity, body consistency, and background stability. Use this s...
Usage Guidance
This skill appears coherent and focused on scoring virtual try-on images. Before installing, consider privacy: the skill will process user images, so avoid uploading sensitive personally-identifying photos (IDs, minors, medical data). Confirm the full SKILL.md (the provided excerpt was truncated) contains no instructions to upload images to external URLs or to read local files outside the user-provided images. If you plan to allow the agent to invoke skills automatically, decide whether you want it to run this scorer whenever images are uploaded; otherwise keep it user-invocable only.
Capability Analysis
Type: OpenClaw Skill Name: virtual-tryon-scorer Version: 1.0.0 The skill bundle contains only metadata and markdown instructions (SKILL.md) for an AI agent to evaluate virtual try-on (VTON) image results. There is no executable code, no network requests, and no instructions that attempt to exfiltrate data or bypass security constraints. The logic is entirely focused on the stated purpose of scoring image quality based on identity preservation and garment fidelity.
Capability Assessment
Purpose & Capability
The name and description match the runtime instructions: the SKILL.md explains how to accept user images, identify panels, and score per-dimension (identity, garment fidelity, body consistency, background). No unrelated binaries, credentials, or config paths are requested, which is proportional to an instruction-only image-evaluation skill.
Instruction Scope
The instructions are focused on analyzing user-provided images, asking clarifying questions when panels are ambiguous, and producing structured scores and explanations. There are no directives in the provided excerpt to read unrelated system files, access environment variables, call external endpoints, or exfiltrate data. (Recommendation: review the remainder of SKILL.md to confirm there are no hidden steps beyond the truncated portion.)
Install Mechanism
There is no install specification and no code files — the skill is instruction-only, so nothing is written to disk and no external packages are pulled in.
Credentials
The skill declares no required environment variables, credentials, or config paths. For the stated task (visual scoring of uploaded try-on images) this is appropriate.
Persistence & Privilege
always is false (not force-included) and model invocation is allowed (the platform default). There is no request for permanent presence or modification of other skills. Autonomous invocation combined with the other benign indicators does not raise concern here.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install virtual-tryon-scorer
  3. After installation, invoke the skill by name or use /virtual-tryon-scorer
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
virtual-tryon-scorer 1.0.0 - Initial release of the virtual try-on quality assessment skill. - Evaluates AI-generated virtual try-on results based on four weighted criteria: face identity preservation, garment fidelity & fit, non-face body identity preservation, and background preservation. - Supports both separate and concatenated image input formats, with logic for identifying image roles and prompting user clarification if needed. - Provides detailed, structured scoring (0–100 per dimension), weighted total score, and concise explanations for each aspect. - Triggers automatically when users request virtual try-on evaluation, scoring, or feedback, including common English and Chinese phrases.
Metadata
Slug virtual-tryon-scorer
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is virtual-tryon-scorer?

Score and evaluate virtual try-on (VTON) results by analyzing identity preservation, garment fidelity, body consistency, and background stability. Use this s... It is an AI Agent Skill for Claude Code / OpenClaw, with 113 downloads so far.

How do I install virtual-tryon-scorer?

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

Is virtual-tryon-scorer free?

Yes, virtual-tryon-scorer is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does virtual-tryon-scorer support?

virtual-tryon-scorer is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created virtual-tryon-scorer?

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

💬 Comments