ClawHub
Back to skill

Security audit

Imgcraft Bare

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small IP-checking utility that openly contacts httpbin.org to print the user's public IP address.

Install only if you are comfortable with a request to httpbin.org revealing your public IP address and ordinary request metadata. The skill appears narrow and non-persistent, but its description should be read as 'public IP checker' rather than a broader connection diagnostics tool.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Lp1

High
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The script performs an outbound network request to a third-party service but the declared permissions reportedly do not cover network access. This creates a permission mismatch that can hide data egress behavior from users or reviewers and is especially relevant because the skill sends a request that reveals the user's public IP to an external service.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill causes an external network request to a third-party service to reveal the user's public IP address, but the skill description does not clearly warn about that data disclosure before execution. Public IP addresses and connection metadata can be sensitive in some environments, and undisclosed outbound requests reduce informed user consent and can create privacy or compliance issues.

Description-Behavior Mismatch

Medium
Confidence
89% confidence
Finding
The manifest describes two outputs: the public IP address and connection information. The implementation fetches https://httpbin.org/get and logs only data.origin, which is the apparent client IP, without showing any additional connection metadata such as headers, user agent, or network details.

Static analysis

No suspicious patterns detected.