ClawHub
Back to skill

Security audit

Flow Hugo Publisher

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Hugo-to-GitHub-Pages publishing helper; it can change and publish a repository, but that matches its stated purpose.

Install this only if you want an agent to manage and publish a Hugo site. Before approving publish steps, verify the workspace, Git remote, branch, staged files, generated workflow, and any custom theme repository. Remove or edit the local state file if you do not want saved workspace and deployment history reused later.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The document provides ready-to-run shell templates that write files, create commits, and push to a remote repository, which can immediately publish content and CI configuration without any adjacent warning or explicit confirmation step. In the context of an agent skill for managed publishing, this increases the risk of unintended repository modification or remote publication if the commands are surfaced or executed automatically.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
This workflow instructs the agent to stage broad changes, commit them, and push to a remote repository, which can modify repository contents and publish changes externally without an explicit high-friction safety check. In an agent setting, this is dangerous because a mistaken workspace, unexpected file additions, or user confusion could lead to accidental publication, leakage of sensitive files, or irreversible remote changes.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill allows a user-supplied arbitrary Git repository to be fetched as a theme/template with no warning or trust boundary, which introduces third-party code and configuration directly into the workspace. In this context, themes can include layouts, assets, submodules, and build-affecting files that may execute during tooling workflows or silently alter site output, making supply-chain compromise and unsafe code ingestion a real risk.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.