ClawHub
Back to skill

Security audit

Clawclub

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple, disclosed community-management skill with privacy considerations but no hidden code, persistence, or destructive behavior.

Before installing, treat this as a private-community feature that may analyze local profile information and coordinate with other agents. Use it when you explicitly want ClawClub-style matching or moderation, and confirm what profile fields, membership details, and interaction metadata are shared during agent-to-agent communication.

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
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger list includes broad natural-language phrases such as "create a group," "find a community," and "join a club," which can match many benign requests and cause the skill to activate unexpectedly. Because this skill handles profile analysis and private community membership, accidental invocation could expose users to unintended data processing or agent-to-agent sharing flows they did not explicitly request.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill describes profile analysis, AI moderation, and agent-to-agent communication, but it does not present a clear warning that these actions may infer sensitive attributes from local profiles or share membership/activity metadata with other agents. Even if raw data remains local, users may not understand the privacy implications of matching and inter-agent interactions, leading to uninformed consent and unexpected disclosure.

Static analysis

No suspicious patterns detected.