avatar-generator

• Outcome: Pass
• Latest audit: May 28, 2026, 11:41 AM
• Version:
  0.1.2

Overview

This skill is a disclosed avatar generator that runs a pinned npm package and writes the resulting SVG file where the user or skill instructions specify.

Install only if you are comfortable with your agent running a pinned npm package through npx and creating SVG files on disk. Prefer specifying an explicit output path, review before overwriting existing files, and use a sandbox or audit the npm package first if your environment is sensitive.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Missing User Warnings

Low

• Confidence: 77% confidence
• Finding: The README instructs the agent to write SVG output to arbitrary user-specified paths, including auto-deriving filenames when a directory is supplied, without any caution about overwriting existing files or restricting destination paths. In an agent context, that can lead to unintended file clobbering or writes into sensitive locations if the path is attacker-controlled or insufficiently reviewed.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The README instructs the agent to save output to a user-specified path, including absolute or relative paths, without guardrails against overwriting existing files or writing outside an allowed workspace. In an agent context, this can lead to clobbering user files, writing into sensitive locations, or enabling path abuse when the model follows untrusted prompts literally.

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The trigger description is overly broad, including generic phrases like 'generate', 'make', or 'produce an avatar' tied loosely to names or IDs. This can cause the skill to activate in unintended contexts and launch an external `npx` command unexpectedly, increasing the chance of surprise network access and file creation without clear user intent.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The skill writes an SVG to a default path in the current working directory when the user does not provide one, but does not require explicit confirmation before performing the write. In an agent setting, implicit file creation can surprise users, overwrite expected workspace contents, or be chained with broad triggering to create unintended artifacts.

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The trigger text is very broad (`create`, `generate`, `make`, `produce an avatar`, and any mention of avatar generation tied to a deterministic input), which can cause the skill to activate in contexts the user did not intend. In an agent setting, overbroad invocation can lead to unexpected package execution via `npx`, unintended file writes, and unnecessary network/package-install activity from vague avatar-related prompts.

VirusTotal

VirusTotal findings are pending for this skill version.