ClawHub
Back to skill

Security audit

chinese-api-docs

Security checks for vulnerabilities and agentic risk

Overview

This is a Chinese API documentation helper, but its external validation backend is not clearly bounded for sensitive API specs or internal documentation.

Review before installing. Use it for non-confidential Chinese API documentation, and do not send private API specs, source code, internal endpoint details, real auth flows, tokens, or production examples to the validation backend unless you explicitly trust that service and understand what it receives.

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
83% confidence
Finding
The skill advertises an executable `scripts/validate.sh` and an external validation backend, which introduces shell-capable behavior without any declared permission model or user-consent boundary. In an agent setting, undeclared execution capability increases the risk of unexpected command execution or network activity, especially because the skill content encourages running a local script.

Tp4

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding
The skill description promises documentation generation and API-backed validation, but the analyzed behavior reportedly only contacts a fixed remote endpoint for health checks and lacks the claimed core functionality. This mismatch is dangerous because users may trust the skill with sensitive API specs or internal documentation under false pretenses, while the skill may instead send data externally or fail to perform the expected validation safeguards.

Static analysis

No suspicious patterns detected.