ClawHub
Back to skill

Security audit

Skill Section Fixer

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed SKILL.md repair tool, but it can rewrite many installed skills at once without built-in confirmation or rollback.

Review before installing. Use it only on a backed-up, intended skills directory; run dry-run first, inspect the planned file count and changes, and avoid broad batch repair unless you are comfortable with many SKILL.md files being rewritten.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding
The skill explicitly instructs users to run a Python script that reads and writes SKILL.md files, and even supports batch modification of an entire skills directory, yet it declares no permissions or equivalent safety boundaries. This mismatch is risky because users and orchestration systems may treat the skill as low-risk while it can modify many local files, increasing the chance of unintended bulk changes or abuse if pointed at the wrong path.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The example trigger phrase, such as '批量修复所有技能', is broad and maps to a high-impact action that can modify a large number of files. Because the skill performs write operations in batch mode, an overly general conversational trigger increases the likelihood of accidental invocation or unsafe automation without sufficiently specific user intent and scope confirmation.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.