ClawHub
Back to skill

Security audit

绿灵·Blooming Elf-v4

Security checks across malware telemetry and agentic risk

Overview

This plant-care skill persistently stores local plant records and reminder state, but that behavior is disclosed, scoped, and aligned with its purpose.

Install only if you want the agent to maintain a local plant database, logs, backups, and optional reminders under ~/.workbuddy. Use the authorization flow only for the intended plants.json, and review scheduled reminders before confirming them.

SkillSpector

By NVIDIA
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
97% confidence
Finding
The skill explicitly instructs the agent to read and write user files under ~/.workbuddy and to invoke shell commands such as cp and python3 scripts/commit_state.py, yet the skill metadata declares no permissions. This creates a hidden capability/permission mismatch: a reviewer or platform may believe the skill is conversational-only, while it actually performs persistent file access and shell execution on user data.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.