ClawHub
Back to skill

Security audit

Geo Social Publish

Security checks for vulnerabilities and agentic risk

Overview

This deprecated skill is not clearly malicious, but it gives agents actionable steps for authenticated publishing despite telling users not to use it.

Install only if you intentionally need this legacy publishing workflow and are prepared to supervise every authenticated API call and publish command. Treat any local API key as sensitive, confirm what content will be exported or uploaded, and prefer the supported replacement process if one exists.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Description-Behavior Mismatch

Medium
Confidence
88% confidence
Finding
The skill presents itself as deprecated and tells users not to install it, but then supplies complete operational instructions for installing tooling, retrieving authenticated content from a remote API, publishing through a local CLI, and reporting results back. This contradiction creates a social-engineering risk: agents or users may still execute the documented workflow, exposing credentials and transmitting content despite the deprecation notice.

Intent-Code Divergence

Medium
Confidence
90% confidence
Finding
The document explicitly says not to execute the deprecated publishing path, yet immediately provides step-by-step instructions to do exactly that, including login, PATH setup, API export, upload, and status reporting. In an agent context, conflicting instructions are dangerous because the procedural content is more likely to be followed than the warning, leading to unintended external actions under user credentials.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The examples show reading an API key from local files and using it in authenticated requests that export content and report publishing status, but they do not prominently warn that this sends authenticated data to external services. In a skill consumed by agents, this omission can normalize credential use and data transmission without adequate user awareness or consent checks.

Static analysis

No suspicious patterns detected.