ClawHub
Back to skill

Security audit

灵感素材整理器

Security checks for vulnerabilities and agentic risk

Overview

This appears to be a legitimate code-review helper, but it defaults to running nested review tools with very broad local authority.

Install only if you are comfortable with a review helper that can run nested Codex in full-access mode by default and may fall back to other AI reviewer CLIs with your diff. Prefer using --no-yolo or AUTOREVIEW_YOLO=0, set fallback reviewer behavior explicitly, and avoid running it on repositories containing secrets or private code you do not want sent to external tools.

SkillSpector

Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill instructs use of a Python script that reads environment variables and performs local file reads/writes, but the skill metadata declares no permissions. This creates a transparency and least-privilege problem: an agent or reviewer may invoke the skill without realizing it can access or modify local data paths, including paths overridden through environment variables.

Static analysis

No suspicious patterns detected.