Security audit

zeelin-twitter-x-auto-ops

Security checks for vulnerabilities and agentic risk

Overview

This Twitter/X automation skill can publish from a logged-in account automatically and relies on hardcoded external helpers without clear approval or account scoping.

Review this skill carefully before installing. Use it only with an X account you intentionally want automated, inspect the external helper scripts it calls, and require manual review before any post, reply, quote, comment, or follow action. Be aware that some workflows can promote a fixed ZeeLin report URL from your account.

SkillSpector

Vulnerability Patterns

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

Findings (8)

Intent-Code Divergence

Medium

Confidence: 91% confidence
Finding: The script is presented as a general 'self-ops workflow', but it includes a capability to publish a tweet automatically when the user supplies text. That mismatch matters because users may reasonably expect only passive discovery and browsing steps, while the script can trigger an external account action on X, increasing the chance of unintended posting or social-media abuse.

Vague Triggers

Medium

Confidence: 87% confidence
Finding: The skill advertises broad activation phrases like operating or managing an X/Twitter account, which can cause the agent to invoke account-affecting automation on loosely related requests. In this context, the skill includes posting, commenting, and follow-back behavior, so overbroad triggering increases the chance of unintended social-media actions on a real account.

Vague Triggers

Medium

Confidence: 84% confidence
Finding: The manual activation examples are vague and action-oriented without clear scope constraints, making accidental invocation more likely. Because this skill is designed to perform externally visible actions on a live social account, ambiguous triggers can translate into unintended posting or engagement activity.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The skill explicitly describes automatic posting and unattended browser publishing, but does not pair those capabilities with a clear warning that they affect a user's public account. In a social-media automation context, unattended posting can create reputational damage, policy violations, or unauthorized actions if triggered incorrectly or used with the wrong account session.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The one-command quick mode collapses discovery, generation, and posting into a single flow that can publish content without an explicit safety checkpoint. This is especially risky because the resulting content is public and may be inaccurate, unwanted, or posted from the wrong authenticated session.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The script performs a remote network fetch and then automatically posts content to a social platform without any user confirmation, preview, or clear warning. In a skill context, this can cause unintended external actions, reputational harm, or abuse of authenticated posting sessions if invoked unexpectedly or chained by another workflow.

Missing User Warnings

Low

Confidence: 88% confidence
Finding: The script opens an external network URL automatically based on user-supplied input, which causes an outbound request and launches a browser without any notice or confirmation. While this is likely intended behavior for a trend-search helper, unencoded input is inserted directly into the query string and the implicit network action may surprise users or trigger unwanted browsing to a malformed or manipulated URL.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: This wrapper executes several helper scripts with no warning or summary of what external actions they may perform, so a user cannot make an informed decision before running them. In a social-media automation skill, hidden helper behavior is especially risky because those scripts could open browsers, scrape data, interact with accounts, or perform posting/follow/comment actions under the user's credentials.

Static analysis

No suspicious patterns detected.