MotuArt Color Engine

• Outcome: Review
• Latest audit: Aug 12, 2026 · 7:23 AM UTC
• Version: 1.0.9

Overview

This skill appears purpose-built for MotuArt image processing, but it deserves review because it can send sensitive portrait and ID-photo images to a remote API through broad or implicit invocation without a runtime consent prompt.

Install only if you are comfortable sending portraits, headshots, and possible passport/ID-photo images to MotuArt's remote service using an API key. Before running it on identity documents or biometric images, confirm the endpoint, account, privacy/retention expectations, and credit costs; prefer explicit invocation and user confirmation before any upload.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (7)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 96% confidence
• Finding: The skill instructs use of local shell scripts, file reads/writes, and networked API calls, but there is no declared permissions model indicating those capabilities. That gap can prevent users or the host platform from understanding that local files may be accessed and portrait/ID images may be transmitted and written to disk, increasing the risk of unintended data exposure or over-privileged execution.

Vague Triggers

Medium

• Confidence: 87% confidence
• Finding: The trigger list is very broad and includes common phrases such as portrait grading, skin tone, LinkedIn photo, background swap, and generic Chinese equivalents. Overbroad triggers can cause the skill to activate unintentionally in unrelated contexts, leading to accidental shell execution, file handling, or transmission of sensitive portrait/ID-photo data to the hosted API.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The skill processes portraits, headshots, and ID/passport photos through a hosted HTTP API, but the description does not prominently warn users that their images will leave the local environment. Because these images are highly sensitive biometric and identity-related data, lack of upfront disclosure undermines informed consent and can result in accidental sharing of personal data with a third party.

Vague Triggers

Medium

• Confidence: 92% confidence
• Finding: The skill enables implicit invocation without any trigger constraints, exclusions, or narrowing conditions, which increases the chance the agent will automatically route portrait or ID-photo related requests to this external skill without clear user intent. Because this skill appears to support sensitive image transformations such as ID/passport photo generation, clothing replacement, and background swaps, unintended invocation could expose personal images or cause high-trust edits to be performed in contexts where they were not explicitly requested.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The script uploads the local input portrait to a third-party remote API endpoint with curl and provides no explicit runtime warning, confirmation, or privacy notice. Because this skill handles portraits, ID photos, passports, and biometric-like personal images, silent exfiltration of sensitive images to an external service is materially risky even if it is the feature's intended behavior.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The script uploads an ID photo and optional report metadata to a remote API endpoint, which can contain highly sensitive personal data. Although this is core to the skill's purpose, there is no explicit user-facing consent, privacy notice, or minimization control in the script, so users may transmit regulated identity data without clear warning or policy enforcement.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The script unconditionally sends the local input file to a remote service via curl multipart upload (`-F "file=@$IN"`) and does not present any explicit warning, consent gate, or locality notice at execution time. In a skill that may be invoked on sensitive portraits, ID photos, passport images, and identity references, this can lead to unintended transmission of highly sensitive personal data off-system if users assume processing is local.

VirusTotal

VirusTotal engine telemetry is currently stale for this artifact.

Static analysis

No suspicious patterns detected.