Back to skill

Security audit

今日话题解读器

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward Chinese trending-news briefing helper that uses web lookups as expected and does not request local access, persistence, or code execution.

Install this if you want Chinese-language, live hot-topic briefings from public web sources. Be aware it may browse/search when you ask broad current-events questions, and you may need to explicitly request another language if you do not want Chinese output.

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 (3)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
83% confidence
Finding

The skill is written entirely as Chinese-first and implicitly steers operation and output into Chinese without documenting language selection behavior. This can override user language preferences or produce inaccessible output, which is a safety and usability concern because it may cause miscommunication in high-context news analysis tasks.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill description is broadly framed around common requests like '今天有什么大事' and '热点解读', which can cause the agent to invoke the skill for ordinary conversational queries without strong routing boundaries. This increases the chance of unnecessary web access, over-collection of external content, and unintended substitution of the base assistant's safer default behavior.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The model-facing trigger list and workflow are ambiguous and permissive, telling the model to use the skill for loosely defined requests about major events or topic discussion. In practice, this can lead to over-triggering, causing the agent to browse external sources and generate authoritative-seeming summaries when the user may not have requested live retrieval or when simpler in-model responses would suffice.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.