ClawHub
Back to skill

Security audit

Visual DNA

Security checks for vulnerabilities and agentic risk

Overview

The verdict is benign with medium confidence: scanners found no AIG issues and judged the reported concerns compatible with a visual design tool, while one referenced path was missing.

The verdict is benign, not risk-free. Before installing, consider that the scan was partial: a referenced path was missing, and Skillspector could not establish the behavior of the skill installed by the CLI command. Treat any extracted design data as potentially nonpublic before committing or sharing it.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The supplied chunk performs skill installation, a distinct environment-changing action not disclosed in the description. It contains no extraction, JSON creation, or UI-generation logic. The behavior of the installed skill cannot be established from this command alone.

Content

No source excerpt is available for this finding.

Instructions to commit and share extracted design data omit a privacy warning.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The skill directs the agent to commit extracted JSON to version control, share it across teams and projects, and feed it to other agents. These actions can persist or broaden access to nonpublic design information, but the file does not warn about those consequences; its external-host disclosure concerns fetching references rather than distributing extracted data.

Content

No source excerpt is available for this finding.