Security audit

Pet Memorial Portrait Generator

Security checks across malware telemetry and agentic risk

Overview

This skill appears to do what it says: it generates pet memorial images by sending your prompt and API token to an outside image service.

Install only if you are comfortable sending the prompt, optional reference image identifier, and Neta/TalesOfAI token to api.talesofai.com. Avoid highly private memorial details, and prefer a limited or disposable API token because the documented usage passes the token on the command line.

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 (6)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 89% confidence
Finding: The skill declares only `tools: Bash` while the documented usage clearly invokes an external API with a token, implying network access. Hidden or undeclared network capability is security-relevant because it can transmit user prompts, metadata, and credentials to third-party services without clear permission boundaries or informed consent.

Description-Behavior Mismatch

Medium

Confidence: 98% confidence
Finding: The skill metadata says it uses the Neta AI image generation API, but the code sends requests and credentials to api.talesofai.com with a different platform identifier. This is a meaningful integrity and transparency issue because users may provide prompts, reference IDs, and API tokens under false assumptions about the receiving service, increasing risk of credential misuse and undisclosed third-party data transfer.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The README tells users to provide prompts, tokens, and optional reference-image identifiers to a third-party image generation API, but it does not clearly disclose that user content and credentials are transmitted to an external service. This creates a privacy and data-handling risk because users may unknowingly send sensitive memorial descriptions, images, or account tokens off-platform without informed consent.

Vague Triggers

Medium

Confidence: 77% confidence
Finding: The activation guidance says to use the skill whenever someone asks to generate pet memorial portraits, but it does not state boundaries such as requiring explicit user consent for external API use, token availability, or avoiding invocation when sensitive images or personal data are involved. Overbroad invocation criteria can cause the agent to call the skill in situations where the user did not expect third-party data transfer.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The documentation instructs users to pass the API token directly on the command line, which can expose credentials through shell history, process listings, logs, CI output, or telemetry. This is a common credential-handling weakness that can lead to token theft and unauthorized use of the external API.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The code transmits the user prompt, optional reference identifier, and authentication token to an external API, but the skill description does not clearly disclose that this content leaves the local environment or that a different service than advertised receives it. In a memorial portrait context, prompts and reference IDs may include sensitive personal or emotional content, making undisclosed transmission more privacy-impacting.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.