Back to skill

Security audit

X (Twitter) Hot Topics

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed X/Twitter trend-fetching helper with purpose-aligned browser scraping and an optional read-only API fallback, with no evidence of hidden execution, persistence, or destructive behavior.

Install only if you are comfortable letting the agent use an already logged-in X browser session to read trends and search results. Set `X_BEARER_TOKEN` only if you want the API fallback, and treat fallback results as partial, especially because the script filters to English.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (4)

Tp4

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The declared behavior materially differs from what the skill can actually do, while also omitting disclosure that it accesses an environment-stored bearer token. This can mislead operators into granting or running the skill under false assumptions, causing overtrust, unexpected data access, and unsafe automation decisions based on incomplete or fabricated capability expectations.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding

The skill uses an environment credential (X_BEARER_TOKEN) but declares no tool scope or permission boundary, so consumers cannot tell that it may access secrets or external APIs. This increases the chance of unintended credential use and weakens least-privilege controls, especially in agent platforms that rely on manifest-declared permissions for policy enforcement.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
79% confidence
Finding

The markdown explicitly recommends using the lang:xx search operator, which constrains results by language. The file does not state that language filtering should be based on user preference or provide an opt-in/choice mechanism, so it can impose a locale policy without consent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The search query appends lang:en unconditionally, which restricts results to English regardless of user preference or context. This is a natural-language policy concern because it enforces a locale/language choice without offering an opt-in or alternative.

Content

No source excerpt is available for this finding.