ClawHub
Back to skill

Security audit

academic-literature-ppt

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed academic-paper-to-Chinese-PPT workflow with no executable code, persistence, credential use, or hidden data movement found.

Install this if you specifically want academic literature turned into a Chinese-language PPT deck. Users who want English, bilingual, or non-PPT output should give explicit instructions or choose a different skill; no evidence of hidden execution, exfiltration, persistence, or credential access was found.

A.I.G

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

SkillSpector

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 (4)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger description is broad enough to match many generic requests about summaries, reports, slides, or article-to-presentation conversion, which can cause the skill to activate when the user did not explicitly ask for this specialized workflow. Unintended invocation is risky here because the skill also imposes strong output constraints, including PPTX production and Chinese translation, so misrouting can produce the wrong artifact or override user expectations.

Natural-Language Policy Violations

Medium
Confidence
97% confidence
Finding
The skill hard-codes Chinese as the primary output language and states that content is translated into Chinese, without requiring user consent or checking the user's language preference. If invoked unintentionally or used in multilingual contexts, this can cause unauthorized transformation of user-requested output, reduce usability, and create integrity issues where the output no longer matches the user's explicit format or language needs.

Natural-Language Policy Violations

Medium
Confidence
95% confidence
Finding
The contract hard-requires Chinese section names and translated output without indicating that language should be user-selectable or constrained by an explicit product requirement. This can cause the agent to ignore a user's requested language, reducing user control and potentially leading to miscommunication or unusable output, but it is not a direct security compromise.

Natural-Language Policy Violations

Medium
Confidence
90% confidence
Finding
The document specifies slide titles such as "综述 (1)" and "讨论 (1)" and elsewhere requires bilingual or Chinese-presented content like Chinese section labels and translating figure/table labels. This imposes a specific language/locale in the skill instructions without user opt-in or a clearly documented region-specific justification.

Static analysis

No suspicious patterns detected.