Security audit

Deal Genie Enhanced

Security checks across malware telemetry and agentic risk

Overview

This markdown-only e-commerce helper is coherent and purpose-aligned; its main risk is optional sharing of business reports to Feishu after the user provides a webhook.

Before enabling Feishu, confirm the webhook points to the intended group or document and that the audience is allowed to see product, sales, ROI, customer-feedback, and influencer-operation details. Avoid pasting regulated, confidential, or personal data unless you are comfortable with that data being included in generated reports or pushed to Feishu.

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 (5)

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The trigger phrases are broad, conversational, and map to large portions of the workflow, which increases the chance of accidental or implicit activation during ordinary chat. In a skill that can generate operational plans and optionally prepare external notifications, unintended invocation can cause oversharing of business content, user confusion, or downstream execution of actions the user did not clearly request.

Vague Triggers

Medium

Confidence: 89% confidence
Finding: Allowing arbitrary single-step triggering without clear boundaries makes it easier for partial user phrases to invoke workflow components out of context. This is especially risky because some steps produce persuasive content, outreach scripts, or execution plans that may be generated without the prerequisite compliance and risk checks described earlier in the skill.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The Feishu integration states that results will be automatically pushed after configuration, but it does not prominently warn users that analysis results, operational schedules, and performance data will be transmitted to an external service. If users provide sensitive commercial information, customer data, or internal performance metrics, this can lead to unintended external disclosure and persistent data exposure outside the immediate chat context.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The document encourages users to provide a Feishu webhook and states that analysis results, push reminders, follow-up prompts, and review reports will be automatically sent to a Feishu group or document. While it mentions the webhook is only used for sending and includes a brief privacy note, it does not clearly foreground that potentially sensitive business, sales, operational, and performance data will be transmitted to an external collaboration space that may have broader audience visibility than the user expects.

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The default trigger phrase “帮我出素材” is broad and likely to appear in ordinary conversation, which can cause the skill to activate without clear user intent or sufficient task scoping. In this skill’s context, unintended activation could generate marketing assets, claims, and promotional content for the wrong product or scenario, increasing the risk of inappropriate or non-compliant output.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.