ClawHub
Back to skill

Security audit

weread-ai-brain

Security checks across malware telemetry and agentic risk

Overview

This skill coherently supports WeRead analysis using a disclosed API key and fixed official gateway, with user confirmation gates for private reads and exports.

Install only if you are comfortable giving the skill a WeRead API key and letting it read the WeRead data you request. Review the data range before approving each fetch or export, and keep generated reports in locations appropriate for private reading notes.

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
88% confidence
Finding
The skill declares no explicit permissions, yet its instructions require shell-capable operations via local scripts and binaries such as curl and python3. This creates a permission-model mismatch: the skill can access private WeRead data and write local exports through shell execution without a clearly declared, enforceable capability boundary, increasing the risk of overbroad data access or unsafe command usage if the skill is invoked improperly or modified later.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.