Back to skill

Security audit

Qiuqiu Helper

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple workspace helper, but it includes persistent file writes and log deletion with too little scoping or confirmation guidance.

Install only if you are comfortable with a helper that can append persistent notes and delete old log files. Before using it, require explicit confirmation for writes and deletions, restrict log cleanup to a known logs directory, and avoid saving sensitive information through quick_note.

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
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The clean_logs tool performs destructive deletion of files, yet the markdown provides no warning about deletion risk, no safety constraints, and allows an optional path parameter that could broaden the deletion target. In a broadly scoped helper skill, this creates a meaningful risk of accidental or unsafe data loss if invoked with an unintended directory or vague user request.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill is described as a 'multi-purpose helper' with broad invocation guidance, which makes its activation boundary unclear and increases the chance it will be invoked in contexts the user did not specifically intend. Because the skill includes file-writing and file-deletion capabilities, ambiguous scope can lead to unnecessary or unsafe tool use without clear user awareness.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The quick_note tool writes content to files in memory, but the description does not warn users that it modifies persistent state or clarify where data is stored. Without explicit notice and consent expectations, the skill may append sensitive, incorrect, or unwanted content to files, creating integrity and privacy risks.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.