Back to skill

Security audit

AI日报测试版

Security checks for vulnerabilities and agentic risk

Overview

This is a simple AI news briefing skill that asks the agent to search public web sources and format a structured daily summary, with no hidden code or persistence.

Install this if you want an AI-news daily briefing workflow. Expect it to perform web searches and produce Chinese-oriented Markdown summaries; users who need another language or a narrower trigger should adjust those instructions before use.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The trigger text is intentionally broad enough to auto-activate on many loosely related requests about AI updates or summaries, which can cause the agent to invoke this skill when the user did not explicitly ask for it. Overbroad routing increases the chance of mis-selection, unintended web access, and the model being steered into a predefined workflow instead of following the user’s actual intent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill description strongly biases output toward Chinese phrasing and terminology without checking the user’s preferred language, which can override user expectations and reduce transparency in multilingual settings. While not a classic security flaw, this creates prompt-level policy misalignment that can lead to incorrect handling of user requests and degraded trust or usability.

Content

No source excerpt is available for this finding.