ClawHub
Back to skill

Security audit

MedXpert 文档模板工具链

Security checks for vulnerabilities and agentic risk

Overview

The skill is a disclosed medical-document toolchain, but it asks agents to handle approval stamps, electronic signatures, trace watermarks, and broad document triggers without enough scoping or user-control detail.

Review this skill before installing in regulated or enterprise document workflows. Only use it when you intentionally want the MedXpert medical-device document chain, and require human confirmation before adding approval stamps, signatures, dynamic user/account watermarks, unlock controls, or billing-related gates to documents.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill metadata declares many generic trigger phrases such as document template, PDF export, approval flow, e-signature, and bilingual translation, which can cause the skill to activate in contexts broader than the user intended. Over-broad invocation increases the chance that sensitive document-processing behaviors are applied unexpectedly, especially in enterprise or regulated-document workflows.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The markdown trigger section repeats a wide set of broad keywords and says the skill should load whenever users mention common document tasks. In practice, this can hijack unrelated sessions and route users into a workflow that includes watermarking, traceability, approval, and signature features they did not request.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill advertises dynamic tracing watermarks, approval stamping, and electronic signature injection, but it does not mention consent, audit visibility, retention, or privacy warnings. These operations can embed user identifiers or alter document trust signals, creating privacy, compliance, and integrity risks if used without explicit user awareness.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
64% confidence
Finding

The skill is primarily written as Chinese-first instructions and labels, while also advertising bilingual output, but it does not tell users they can choose their preferred operating language for prompts or workflow guidance. This may create a locale constraint by default rather than presenting language as an explicit user option.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.