ClawHub
Back to skill

Security audit

Howtoletmyagent Installer

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed installer skill that can add persistent companion skills, but it requires source checks and user approval before writes or install commands.

Install only if you trust howtoletmyagent.xyz and the specific manifest being fetched. Before approving, review the exact URL, manifest file list, target install path, and any install command; do not approve unexpected domains, shell commands, or paths outside the intended skills folder.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
82% confidence
Finding
The skill allows activation from a vague request like "learn the skill from this page" based only on URL context, which can cause the agent to act on unintended or attacker-supplied links present earlier in the conversation. In an installer skill, broad invocation increases the chance of fetching and installing a manifest without sufficiently clear user intent tied to a specific source.

Static analysis

No suspicious patterns detected.