Back to skill

Security audit

Daily Inspiration

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Chinese-language daily inspiration push workflow, with some incomplete implementation and configuration details but no evidence of hidden, destructive, or credential-stealing behavior.

Install only if you want Chinese-language daily inspirational content and are comfortable granting the agent permission to search the web and send scheduled messages through Feishu or the platform message tool. Confirm how the schedule, recipient, and disable controls are configured, because the artifact does not define those setup details.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

声明描述的是一个会定时向用户推送精神修行相关资讯的技能,且明确提到通过飞书发送、内容动态来源于相关大师主题。实际代码只包含一个主题词库、按早晚随机选题、简单消息格式化函数,以及主程序中打印所选主题和时段。虽然主题方向与声明一致,但核心能力明显不符:没有任何飞书接口调用、没有外部内容抓取或生成、没有真正发送给用户,也没有调度逻辑。故描述与实际行为存在实质性不匹配。

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill is designed to run automatically on a schedule and send outbound Feishu messages without any visible user-facing warning, consent flow, or configuration notice in the skill description. This is dangerous because it can lead to unexpected automated messaging, privacy issues, or misuse of a connected communication channel even if the content itself is low risk.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file’s docstring describes the skill entirely in Chinese, and the script also prints Chinese-only user-facing output. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale constraint is explicitly documented and justified.

Content

No source excerpt is available for this finding.