Multi-Platform Poster

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed social-media publishing helper, but users should treat connected account credentials and public posting controls carefully.

Install only if you are comfortable connecting social media accounts and potentially publishing public content. Configure only needed platforms, keep API secrets out of chat, review each adapted post and schedule before confirming, and separately review any dependent publishing, rewriting, or image-handling skills.

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 (7)

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The example trigger phrases are very broad imperative phrases like '发布到微信和知乎' and '多平台发布这篇文章', which can plausibly appear in normal conversation, drafting, or quoted text. In a skill that can post to external accounts, ambiguous activation raises the risk of unintended publishing or scheduling actions without sufficiently explicit user intent.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The README states that after entering Markdown, the user can simply say '发布到所有平台', but it does not define scope checks, draft-vs-send behavior, or confirmation requirements. Because this skill performs real actions against external social accounts, ambiguous activation materially increases the chance of accidental mass posting or unintended scheduled content distribution.

Missing User Warnings

High

Confidence: 95% confidence
Finding: The skill advertises one-click publishing and scheduled posting to multiple external platforms but provides no clear warning that these actions modify external accounts and may distribute user content irreversibly. Users may not understand that publishing, timing, and later deletion are consequential account actions, increasing the risk of accidental posting, reputational harm, or loss of control over distributed content.

Missing User Warnings

Medium

Confidence: 87% confidence
Finding: The README describes a cross-platform performance dashboard with reads, likes, comments, and shares, but does not warn users that the skill may collect, aggregate, store, and display account analytics data. This omission creates privacy and data-governance risk, especially if the data includes account-level metrics or content performance that should be retained minimally and disclosed clearly.

Missing User Warnings

High

Confidence: 97% confidence
Finding: The authorization section instructs users to bind platform accounts and provide sensitive credentials such as AppID, AppSecret, App Key, and Client Secret without any warning about credential sensitivity or the level of access being granted. This is dangerous because it normalizes entering high-value secrets into the skill and may lead to credential compromise, overbroad account access, or unauthorized posting if storage and handling are not tightly controlled.

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The trigger phrases are broad and generic for social media publishing, which increases the chance of accidental or context-insensitive activation. In a skill capable of cross-platform posting, unintended activation could lead to unauthorized publishing, reputational harm, or disclosure of content to external services.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The skill advertises cross-platform distribution and analytics collection but does not clearly disclose what content, account metadata, or engagement data will be shared with third-party platforms. This creates privacy and consent risks because users may not realize their content and performance data will be transmitted, stored, or correlated across multiple services.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal