Back to skill

Security audit

Zhizhi Math Worksheets

Security checks for vulnerabilities and agentic risk

Overview

This skill coherently creates local math worksheets and answer keys, with some expected local storage of worksheet inputs and outputs that users should manage carefully.

Before installing, be aware that generated worksheets, answer keys, source photos, and answer manifests may be saved locally and could contain student work or personal information. Use a suitable output folder, avoid including unnecessary personal data, and delete old worksheet folders when they are no longer needed.

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
87% confidence
Finding

The default prompt references the trigger phrase "$zhizhi-math-worksheet" but does not clarify when the skill should or should not be invoked beyond a general description. It also provides no negative examples or scope constraints, which makes the activation boundary ambiguous in a manifest file.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The file explicitly instructs the system to save the input list, generated content, and answer manifest, which can include student work and potentially personal or educational data. Because there is no accompanying retention limit, minimization rule, consent notice, or privacy control, this creates a real privacy and data-governance risk if the skill stores sensitive learner information longer than necessary or without user awareness.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.