Security audit

Finance Radar Tool Free

Security checks across malware telemetry and agentic risk

Overview

The skill is mainly a finance analysis helper, but its activation wording and declared read/write/exec authority are broader than its stated single-asset finance purpose.

Review this skill carefully before installing. It appears aimed at finance analysis and does not contain bundled code or obvious malicious behavior, but its activation scope should be narrowed to stock/crypto analysis and any write/export behavior should be clearly defined before granting broad read/write/exec access.

SkillSpector

By NVIDIA

Vulnerability Patterns

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
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (3)

Description-Behavior Mismatch

Medium

Confidence: 81% confidence
Finding: The documentation says the free edition only supports single-asset analysis, yet later states that the core execution supports create/query/export operations. That inconsistency can mislead an agent or user into attempting broader data handling than intended, increasing the chance of unauthorized or unexpected file/report export behavior.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The invocation guidance is overly broad, telling agents to use the skill for generic data analysis, report generation, statistical insight, and visualization tasks. In an agent ecosystem, such broad matching can cause the skill to be selected for unrelated prompts, leading to unnecessary execution, file access, or command use beyond the finance-focused scope.

Vague Triggers

High

Confidence: 95% confidence
Finding: The capability coverage section includes excessively broad and fragmented keywords, including generic terms like 'Use', 'when', '数据分析', '报表生成', and '自动化工作流场景'. In a tool-selection pipeline, this can act as prompt-squatting, causing the skill to trigger on many unrelated requests and granting it read/write/exec opportunities in contexts far outside stock analysis.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.