语构 App Builder

Security checks across malware telemetry and agentic risk

Overview

This skill mostly matches its app-builder purpose, but it gives an agent broad authority to publish and change remote apps, including automatic publishing guidance that users should review first.

Install only if you trust the Yugoo service and are comfortable giving this skill an API key that can create, modify, publish, upload to, and delete remote app state. Before use, require explicit user approval for publishing, deleting conversations or versions, uploading local files, cloning/copying projects, or using --insecure; prefer a revocable least-privilege key.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

Confidence: 84% confidence
Finding: The routing keywords are broad enough to capture generic web-development requests, which can cause the agent to invoke a high-impact external builder skill unexpectedly. In context, this skill can create, modify, publish, clone, and delete remote applications, so over-triggering increases the risk of unintended network actions and secret-backed operations.

Missing User Warnings

Medium

Confidence: 87% confidence
Finding: The CLI exposes an --insecure flag that disables TLS certificate verification, which can enable man-in-the-middle attacks against authenticated API traffic if users rely on it. Because this tool handles bearer tokens and deployment actions, disabling certificate validation meaningfully increases the risk of credential theft and tampering.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal