Security audit

律师全案攻防实战推演

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent legal case-strategy skill with a disclosed, user-invoked DOCX export helper, not hidden or automatic data handling.

Installers should treat this as a legal workbench that may process confidential case material if the user chooses to run the export script. Run the converter only on specific intended Markdown files or case directories, and review generated DOCX/ZIP outputs before sharing them.

SkillSpector

Vulnerability Patterns

MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
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 (2)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 84% confidence
Finding: The skill advertises and invokes filesystem-related capability via the DOCX export workflow and references reading templates/resources, but no permissions are declared. Undeclared file-read behavior weakens user and platform visibility into what data the skill may access, which is especially sensitive in a legal-analysis context where case files may contain confidential client material.

Tp4

High

Category: MCP Tool Poisoning
Confidence: 92% confidence
Finding: The skill is described as a legal strategy workbench, but it also includes operational document-conversion behavior such as recursive directory scanning and ZIP packaging. This mismatch can cause users or downstream policy systems to trust the skill with sensitive case data without realizing it may process broad sets of local files, increasing the risk of unintended collection or transformation of confidential documents.

Static analysis

No suspicious patterns detected.