ClawHub

Ai Project Scaffold

Security checks across malware telemetry and agentic risk

Overview

This skill is a project scaffold and maintenance toolkit whose file changes are disclosed and fit its purpose.

Install only if you want a scaffold that can create and update files in a project directory. Use dry-run for checkpoint cleanup first, and be aware that doc_lint --fix edits documents in place and checkpoint_mgmt clean deletes checkpoint files unless --dry-run is used.

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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill clearly instructs the agent to create directories, generate templates, copy scripts, and upgrade existing projects, which requires filesystem read/write behavior. Having those capabilities implied in the content but not explicitly declared weakens permission transparency and can cause an agent runtime to perform broader file operations than a user expects.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
In --fix mode, the script prepends a marker comment directly into matching files, modifying user content in place. Although gated by an explicit CLI flag, the write happens without backup, atomic replace, dry-run preview, or per-file confirmation, so a user can unintentionally alter many documents and potentially corrupt formatting-sensitive files.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal