GitHub Skill Updater

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed GitHub-based skill updater that can change local skills, so it should be used deliberately but does not show hidden or malicious behavior.

Install this only if you want an agent to check and modify local git-cloned skills. Before running updates, confirm the target path and git remote, review upstream changes when possible, and keep a rollback path for any skills that affect important workflows.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Rogue AgentSelf-Modification, Session Persistence
  • 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
91% confidence
Finding
The skill instructs the agent to execute shell commands and perform repository updates that modify local files, yet it declares no permissions. This creates a capability/permission mismatch that can bypass user expectations and safety controls, especially because the update path performs writes via git operations and script execution.

Self-Modification

High
Category
Rogue Agent
Content
更新单个 skill:

```bash
./skills/github-skill-updater/scripts/github-skill-updater update skills/<skill-name>
```

也可以直接调用 Python 主脚本:
Confidence
88% confidence
Finding
update skill

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal