Back to skill

Security audit

AI News Aggregator

Security checks for vulnerabilities and agentic risk

Overview

The verdict is benign with medium confidence: scanners found no issue in the completed scan, but noted undeclared GitHub trend discovery and some workflow choices, while coverage was partial.

The verdict is benign, not a guarantee of safety. Before installing, consider whether the undisclosed GitHub trend feature and automatic Chinese translation suit your needs. The scanners inspected only half of the reported components, and referenced scripts were unavailable for review, so evidence about their behavior is incomplete.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding

The supplied commands expose GitHub project-trend discovery rather than the declared RSS news-processing workflow. Although topically related to AI and technology, this is a distinct aggregation capability and resource not identified in the description. The invoked script's implementation is not supplied, so its internal behavior cannot be verified; no additional behavior is inferred.

Content

No source excerpt is available for this finding.

The workflow automatically translates English news into Chinese without offering a target-language choice.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The processing workflow explicitly selects English-to-Chinese translation regardless of the user's requested language. The same requirement is repeated at L379, but the skill does not provide language opt-in or establish a regional requirement that makes Chinese output necessary.

Content

No source excerpt is available for this finding.

The trigger “今天有什么新闻” (“What's in the news today?”) does not specify the skill's AI/technology scope.

Low
Category
Trigger Abuse
Confidence
93% confidence
Finding

This phrase is a general news request that could concern local events, sports, politics, or other topics outside the stated AI/technology focus. It is explicitly listed as an activation keyword without a domain qualifier or exclusion condition.

Content

No source excerpt is available for this finding.