ClawHub

ht-skills

Security checks across malware telemetry and agentic risk

Overview

The skill largely does what it says, but its bundled config sends token-authenticated API requests to an unencrypted HTTP endpoint, so users should review it before installing.

Install only if you trust the 灏天文库 service and intend to let the agent create, edit, move, retrieve, and upload content in that account. Before use, change server_base_url to an HTTPS endpoint or set HT_SKILL_SERVER_URL to HTTPS, and keep HT_SKILL_TOKEN private because it authorizes remote account actions.

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 (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill exposes sensitive capabilities via `allowed-tools: [bash]` plus documented use of environment variables, local file reads (`config.json`, image/content files), and networked API calls, yet it does not declare corresponding permissions in a structured way. This can mislead policy enforcement or reviewers about the skill's real access level, increasing the chance that secrets such as `HT_SKILL_TOKEN` or local file contents are accessed or transmitted without adequate scrutiny.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal