Back to skill

Security audit

Ghibli Style Skill

Security checks across malware telemetry and agentic risk

Overview

This skill is a small, user-run image generator that sends a prompt and provided API token to the disclosed Neta/TalesOfAI service, with no evidence of hidden persistence or unrelated access.

Before installing, verify you trust the package source and the Neta/TalesOfAI service. Treat prompts and reference image IDs as data sent to a third party, avoid including confidential or personal information, and prefer a limited-use API token that you can revoke.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (10)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill advertises Bash execution and, per the static finding, performs network access without declaring corresponding permissions. Undeclared network capability weakens user trust and reviewability because prompts, tokens, or generated content may be sent off-host without explicit disclosure or permission metadata.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding
The documented behavior says the skill uses the Neta API, but the static analysis indicates it actually contacts api.talesofai.com and supports undocumented reference-based editing features. This mismatch is dangerous because users may provide API tokens and sensitive prompts under false assumptions about where data is sent and what operations the skill can perform.

Description-Behavior Mismatch

Medium
Confidence
97% confidence
Finding
The skill metadata claims it uses the Neta AI API, but the code actually sends prompts and credentials to talesofai.com. This mismatch is dangerous because users may consent to share data with one provider while the skill silently transmits it to another, defeating informed consent and increasing phishing/supply-chain risk.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The README explicitly states that user prompts and optional reference image identifiers are sent to a third-party image generation API, but it does not provide any privacy notice, data-sharing disclosure, or guidance against submitting sensitive content. This is dangerous because users may unknowingly transmit personal, confidential, or proprietary data to an external service, creating privacy, compliance, and data-handling risks.

Vague Triggers

Medium
Confidence
78% confidence
Finding
The invocation guidance is very broad ('use when someone asks to generate or create...') and lacks constraints around consent, content policy, external transmission, or token handling. Over-broad triggers increase the chance the skill is invoked automatically in contexts where user data should not be sent to third-party image APIs.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The documentation omits a warning that user prompts are transmitted to an external image-generation service. In a skill that takes free-form text and an API token, this creates a privacy and data-governance risk because users may unknowingly submit sensitive or proprietary information to a third party.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
Accepting the API token on the command line exposes it to shell history, process listings, logging, and other local observers. The risk is amplified here because the token is then sent to an unexpected third-party endpoint, so credential compromise could lead to unauthorized API use or account abuse.

External Transmission

Medium
Category
Data Exfiltration
Content
}

// --- Submit job ---
const submitRes = await fetch("https://api.talesofai.com/v3/make_image", {
  method: "POST",
  headers: HEADERS,
  body: JSON.stringify(body),
Confidence
88% confidence
Finding
fetch("https://api.talesofai.com/v3/make_image", { method: "POST"

External Transmission

Medium
Category
Data Exfiltration
Content
}

// --- Submit job ---
const submitRes = await fetch("https://api.talesofai.com/v3/make_image", {
  method: "POST",
  headers: HEADERS,
  body: JSON.stringify(body),
Confidence
88% confidence
Finding
https://api.talesofai.com/

External Transmission

Medium
Category
Data Exfiltration
Content
for (let attempt = 1; attempt <= MAX_ATTEMPTS; attempt++) {
  await sleep(POLL_INTERVAL_MS);

  const pollRes = await fetch(`https://api.talesofai.com/v1/artifact/task/${task_uuid}`, {
    headers: HEADERS,
  });
Confidence
86% confidence
Finding
https://api.talesofai.com/

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.