Back to skill

Security audit

DingJi Long Image

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent Chinese higher-math long-image generator, with disclosed local file output and upload for a download link but no hidden code, persistence, credential access, or destructive behavior.

Install this if you want a Chinese higher-math course long-image generator and are comfortable with generated images being saved locally and uploaded to obtain a download link. Avoid putting private student data, proprietary course content, or sensitive names in prompts unless external upload is acceptable.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
83% confidence
Finding

The README is entirely written as a Chinese-only skill for generating Chinese higher-math course long images, and examples/instructions assume Chinese usage throughout. Under the policy rule, forcing a specific language without user opt-in can be a natural-language policy violation unless the locale constraint is explicitly justified or alternatives are offered.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The README explicitly states the generated image will be uploaded to cloud storage and a download link returned, but it does not disclose this as an external data transfer or mention consent, retention, or sensitivity considerations. Because user-provided chapter names, branding/watermarks, and generated instructional content may contain proprietary or personal information, silent upload can expose data outside the local environment.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger phrases are broad enough to match ordinary requests such as '做一张长图' or '章节总结图', which can cause the skill to activate outside the user's clear intent to use this specific capability. That creates a prompt-scope and consent problem: the agent may generate files or invoke downstream tooling when the user only asked generically for an image or summary.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly writes output to a local filesystem path and uploads it via an external file-upload mechanism, but it does not present this as a user-consent or data-handling warning. If the generated content includes sensitive educational materials, names, or user-provided data, users may unknowingly cause local persistence and external transmission.

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

This markdown file presents all instructions and reference material in Chinese, including the title, descriptions, and usage notes. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation unless the locale restriction is explicitly documented and justified.

Content

No source excerpt is available for this finding.