ClawHub
Back to skill

Security audit

Brother.skill

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local memory-and-style simulation tool with disclosed personal-profile storage and no evidence of hidden code, network transmission, or system-level persistence.

Install only if you are comfortable keeping plain-text local profiles about real people. Use explicit requests when creating or using profiles, avoid sensitive private details, and delete a person's folder when you no longer want their data retained.

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
  • Rogue AgentSelf-Modification, Session Persistence
  • 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 (3)

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 38)May include surrounding context.

md
- Download, scrape, or collect media from any platform
- Monitor or track any person's online activity
- Transmit any data to external servers or third parties
- Create deepfakes or identity impersonation

**User responsibilities:**
- All input is your own memories and observations

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger for Profile Building Mode is overly broad: any user message that merely describes a person, story, or speaking style could be interpreted as consent to create or update a persistent profile. That can lead to unintended collection and retention of personal or sensitive information without an explicit confirmation step, especially because the skill also specifies cross-session persistence and timestamped logging.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The Group Chat Simulation trigger is underspecified and may activate on ambiguous requests involving multiple people or prior stored profiles. Because the feature generates in-character responses for several real people at once, accidental activation increases the risk of unwanted impersonation-style outputs, reputational harm, or disclosure of previously stored personal traits from local memory.

Content

No source excerpt is available for this finding.