Back to skill

Security audit

clawlite-design-consult

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed design-consulting workflow that reads limited project context, optionally researches competitors, creates a local preview, and writes design guidance files.

Before installing, understand that using the preview path may open a browser and load fonts from Google, and the final write step may create DESIGN.md and append design guidance to CLAUDE.md. Review those changes before relying on them in future agent sessions.

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill instructs generating a local HTML file, opening it automatically in the browser, and loading Google Fonts from a CDN, but it does not warn the user that this will launch an application and trigger external network requests. This can leak metadata such as IP address, user agent, timing, and possibly project-derived branding/content embedded in the preview, while also causing unexpected side effects on the user's machine. The danger is moderate rather than high because the destination is a mainstream CDN and the feature is design-related, but automatic launch plus silent network access is still unsafe without consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill directs the agent to write DESIGN.md at the project root and append to CLAUDE.md, but it does not require explicit user confirmation immediately before modifying repository files. That creates a real integrity risk: an agent could alter project documentation or steering files based on inferred context, surprising the user or overwriting existing guidance. The risk is elevated slightly because CLAUDE.md can influence future agent behavior, so silent changes may persist beyond the current interaction.

Content

No source excerpt is available for this finding.