ClawHub
Back to skill

Security audit

social

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a documented inbed.ai social-network API integration, but it asks for broad local read/write/command authority and broad activation beyond that purpose.

Install only if you intend to let an agent manage an inbed.ai social profile, including sending messages, swiping, changing relationship status, uploading photos, and storing/using a Bearer token. Keep the token in a credential manager or environment variable, avoid giving the skill unrelated local-file tasks, and review any command it runs because the requested read/write/exec authority is broader than the platform API workflow requires.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Context-Inappropriate Capability

Medium
Confidence
87% confidence
Finding
The skill claims generic file read/write and command execution capabilities that are not necessary for a social-network API integration. Overbroad capabilities increase the attack surface and can let prompt-driven workflows access local data or execute unintended operations unrelated to the documented business function.

Vague Triggers

Medium
Confidence
81% confidence
Finding
The activation guidance is very broad, invoking the skill for general AI model use, intelligent dialogue, agent orchestration, and LLM applications rather than tightly scoping it to social-network operations. Such loose routing criteria can cause the skill to be selected in unrelated contexts where its exec/write capabilities and external API behaviors become unnecessarily exposed.

Static analysis

No suspicious patterns detected.