Skillv1.0.0

ClawScan security

Bikini Contest Photos – Create Contest-Ready Images Online – API-powered · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

SuspiciousApr 14, 2026, 8:32 AM

Verdict: suspicious
Confidence: high
Model: gpt-5-mini
Summary: The skill is technically coherent with its declared API key requirement and endpoints, but the runtime instructions include a default prompt that explicitly directs producing sexualized/undressed images from person photos — a high-risk behavior (ethical, legal, privacy) that users should be warned about.
Guidance: Technically the skill is coherent: it needs only a WESHOP API key and calls openapi.weshop.ai endpoints. However the skill's default instructions explicitly direct creating sexualized/undressed versions of person photos. Before installing or using it: (1) do not upload images of real people without their explicit consent; this can be illegal or ethically wrong; (2) verify the WeShop service terms, content-moderation policy, and local laws about deepfakes/sexual imagery; (3) only provide your API key if you trust openapi.weshop.ai and ensure calls go to that host (the SKILL.md warns about this); (4) consider refusing or editing out the default 'undress' textDescription if you want to avoid generating sexual or non-consensual content. If you need help assessing legal/ethical risk or sanitizing prompts, seek legal/advisory guidance before proceeding.

Review Dimensions

Purpose & Capability: okName/description, declared endpoint host (openapi.weshop.ai), and the single required env var (WESHOP_API_KEY) align: this skill legitimately needs a WeShop API key to call the provider's OpenAPI endpoints.
Instruction Scope: concernSKILL.md stays within the API's domain (openapi.weshop.ai) and tells the agent to upload images and poll runs. However the default textDescription explicitly instructs 'naturally undress and change the outfit into a thin bikini', which directs generation of sexualized/explicit transformations of person photos. That behavior is ethically and legally sensitive (potential non-consensual deepfakes) and should be considered out-of-scope for many users. The doc otherwise does not ask for unrelated files or extra env vars.
Install Mechanism: okNo install spec and no code files — instruction-only skill (lowest install risk). Nothing is downloaded or written to disk by an installer.
Credentials: okOnly a single credential (WESHOP_API_KEY) is required and is the documented primary credential. The SKILL.md explicitly warns not to send the key to domains other than openapi.weshop.ai. No unrelated secrets or config paths are requested.
Persistence & Privilege: okalways is false, no installs, and the skill does not request system-wide configuration changes or persistent elevated privileges.