ClawHub
Back to skill

Security audit

Factory AI Droid

Security checks across malware telemetry and agentic risk

Overview

This skill is a thin wrapper around a powerful third-party coding agent CLI that can change code, deploy, add extensions, and retain session context without enough safety scoping in the skill text.

Install only if you trust the Factory Droid binary and account connection. Verify the binary source/version before use, avoid `--force` unless you explicitly want automatic changes, and require review before commits, PRs, deployments, plugin installs, or MCP server changes. Confirm how Factory scopes organization code access and clears or isolates session memory.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • 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)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill explicitly documents `droid exec --force` and deployment actions like `deploy to fly.io` without any safety guidance, confirmation expectations, or scope limitations. In an agent skill context, this can normalize destructive or production-impacting actions and increase the chance that users or downstream agents invoke code-modifying or live-system changes without adequate review.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.