GrowthCircle.id Provider

Security checks across malware telemetry and agentic risk

Overview

This looks like a real GrowthCircle provider plugin, but it under-discloses image generation and downloads provider-returned URLs without clear limits.

Install only if you intend to trust GrowthCircle.id with prompts and image-generation requests. Treat the current package as needing review because the public description is narrower than the implemented capabilities, and the image result downloader should be constrained before use in sensitive or internal-network environments.

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 Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 86% confidence
Finding: The documented behavior says the plugin is for an OpenAI-compatible model endpoint, but the analyzed behavior also includes image generation, task polling, and downloading generated assets from returned URLs. That mismatch is dangerous because users and reviewers may approve the plugin expecting only standard model inference traffic, while the plugin can make additional network requests and retrieve remote content, expanding the trust boundary and attack surface.

Description-Behavior Mismatch

Medium

Confidence: 92% confidence
Finding: The plugin metadata describes only a model-provider plugin, but the code also registers an image generation provider. This capability mismatch can mislead users and reviewers about what the plugin does, weakening informed consent and security review because an additional provider surface is exposed without clear disclosure.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The manifest clearly configures this skill to send user prompts and image-generation requests to the external endpoint https://ai.growthcircle.id/v1, but the file does not present any user-facing disclosure or warning about third-party data transfer. In an agent/plugin ecosystem, this can lead users to unknowingly transmit sensitive prompts, images, or embedded secrets to an external service, creating a privacy and data-governance risk even if the provider is legitimate.

Missing User Warnings

Medium

Confidence: 80% confidence
Finding: The code automatically fetches arbitrary URLs returned by the remote provider when converting image results into assets. If the provider is malicious, compromised, or tricked into returning attacker-controlled URLs, this can trigger server-side request forgery behavior, unexpected outbound requests, access to internal network resources, or downloading oversized/untrusted content.

VirusTotal

61/61 vendors flagged this plugin as clean.

View on VirusTotal