Back to skill

Security audit

Group Chat Protocol

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple group-chat behavior guide with no executable code, network access, persistence, or hidden data-handling behavior.

Before installing, confirm the group actually wants this assistant to participate and react in chats, and adjust the AEST quiet-hours assumption if the group operates in another timezone. The privacy rule is appropriate, but users should still ensure the host agent does not disclose private memory or DM context into group conversations.

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
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Category
Not specified by scanner
Confidence
90% confidence
Finding

The activation guidance relies on subjective phrases like 'be smart about when to contribute', 'add clear value', and 'something witty and appropriate fits naturally' without clear constraints or negative examples. In a markdown skill file, this can cause unintended or inconsistent invocation because the boundaries for when the skill should speak are not specific enough.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

The instruction 'Late night (23:00–08:00 AEST) unless urgent' imposes an AEST-specific locale policy on behavior. This is a natural-language policy concern because it forces a specific timezone/locale assumption without offering user choice or documenting why that locale is required.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.