Back to skill

Security audit

Tianshu Paper Card

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple local paper-abstract formatter that reads user-provided text or a chosen file and prints a Chinese Markdown reading card.

Install this if you want Chinese Markdown reading cards from abstracts. Review the generated citation draft against the original paper, and only pass files you intend the local script to read.

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 (3)

Vague Triggers

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

Line L05 lists invocation examples such as 「摘要整理」 and 「论文笔记模板」 without further scope constraints or exclusion examples beyond a general NOT for statement. These phrases are broad enough to match ordinary writing or study assistance requests, which could cause unintended activation of this specific skill.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

The file includes Chinese-facing text in the tagline and tags while the main description is in English. Because the metadata does not state that the skill supports multiple languages or let the user choose a language/locale, it may implicitly steer users into a specific language experience without opt-in.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
97% confidence
Finding

The script's error messages, usage text, placeholders, headings, and output content are written in Chinese throughout, which imposes a specific language on users. There is no opt-in, language selection, or indication that this is intentionally limited to a Chinese-only context.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.