Security audit

IceCube Social Ops

Security checks for vulnerabilities and agentic risk

Overview

This skill is for social-media automation, but it asks for live account/session control and defaults to public automation with too little user control.

Install only if you intentionally want an agent to operate these social accounts. Use an isolated browser profile with only the target accounts logged in, avoid saving reusable cookies, require draft review before posts/replies/follows/DMs, and enable autonomous operation only after you define clear platform, action, and stop conditions.

SkillSpector

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: 89% confidence
Finding: The trigger phrases are broad enough to match common requests about social media, posting, and account operations, which can cause the skill to activate in situations where the user did not intend full account automation. In this skill's context, that is more dangerous because activation can lead to account-impacting actions such as posting, interacting, and handling messages on behalf of the user.

Missing User Warnings

High

Confidence: 97% confidence
Finding: The skill advertises end-to-end automatic login-dependent posting, interaction, and operation of social accounts without a prominent warning that it may take public actions, access private content, and affect account reputation. In context, this is especially risky because the skill is designed for autonomous operation after one-time authorization, increasing the chance of unintended or misunderstood account activity.

Missing User Warnings

High

Confidence: 98% confidence
Finding: The instruction to keep Chrome open or save cookies describes persistence of authenticated sessions without warning that cookies and browser sessions are sensitive credentials. If mishandled, those sessions could let another process or person act as the user on X/Twitter or 小红书 without reauthentication.

Missing User Warnings

High

Confidence: 99% confidence
Finding: The skill sets automatic publishing as the default and treats human approval as optional, which creates a substantial risk of unintended public posts, policy violations, and brand damage. This is more dangerous in a social-media automation skill because publication is immediate, public, and potentially irreversible once posted or amplified.

Missing User Warnings

High

Confidence: 98% confidence
Finding: The skill describes continuous monitoring and handling of comments and private messages without a clear privacy warning or consent boundaries. In this context, that exposes sensitive communications and can result in unauthorized reading, replying, or escalation of private user conversations, with privacy, trust, and compliance consequences.

Static analysis

No suspicious patterns detected.