ClawHub
Back to skill

Security audit

Cosdesign Standalone

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed website design-analysis prompt generator with ordinary privacy cautions around fetching user-provided URLs.

Install only if you are comfortable having an agent fetch and analyze pages you provide. Use public or authorized URLs, avoid internal services and links containing tokens or confidential context, and monitor or delete data/analysis-history.json if you do not want local analysis history retained.

SkillSpector

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

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill explicitly states it appends analysis history containing URLs, timestamps, and summaries to `data/analysis-history.json` on every run, but does not clearly warn users or offer consent/retention controls. Browsing targets and summaries can reveal sensitive research interests, internal URLs, or confidential project context, creating a privacy and data-retention risk even if no credential theft occurs.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The script generates an agent prompt that directs WebFetch to retrieve arbitrary user-supplied URLs without any validation, allowlist, or warning that external network access will occur. In an agent context, this can enable unintended requests to internal services, metadata endpoints, or sensitive URLs if the downstream agent executes the prompt as written, making this an SSRF-style prompt safety issue.

Static analysis

No suspicious patterns detected.