ClawHub
Back to skill

Security audit

Quant-X v13

Security checks for vulnerabilities and agentic risk

Overview

This skill is a public-market stock analysis script that matches its stated purpose, though it saves a result file to a fixed workspace path without saying so upfront.

Before installing, expect the skill to contact Tencent and Eastmoney market-data endpoints and to save a JSON result at /workspace/v13/analysis_result.json when run. Review or change that output path if you do not want fixed-path persistence or overwrites, and treat the generated trading guidance as research only, not financial advice.

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 (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill advertises execution of a Python script that consumes multiple external market data sources, implying network access and likely local output or cache writes, yet no permissions are declared. This creates a transparency and trust problem: users and hosting platforms cannot accurately assess or constrain the skill’s capabilities, and undeclared network/file behavior increases the risk of unexpected data exfiltration, persistence, or unsafe side effects if the implementation changes or is abused.

Description-Behavior Mismatch

Medium
Confidence
90% confidence
Finding
The script writes analysis output to a fixed local path under /workspace without any user consent, configurability, or validation of whether persistence is appropriate in the execution environment. In agent or shared runtime contexts, even non-sensitive market-analysis output can create unintended data persistence, overwrite existing files, or leak context to later tasks/users.

Missing User Warnings

Low
Confidence
92% confidence
Finding
Writing results to a fixed file path without warning or confirmation is a real safety issue in agent environments because it silently persists artifacts and may overwrite prior files. The danger is contextual rather than directly code-execution-related, but in shared or sandboxed workspaces it can expose prior run data or create unintended cross-task state.

Static analysis

No suspicious patterns detected.