Security audit

groupchat-participant

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small, disclosed group-chat participation helper with a mild autonomy caveat around self-joining or leaving chats.

Install only if you want agents to participate in host-mediated group chats. Consider requiring explicit host or user approval for join/leave actions if your chats may contain sensitive information or need strict participant control.

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 (1)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The skill explicitly allows the agent to 'join or leave the chat on its own initiative' without defining strict authorization, trigger conditions, or policy boundaries. That creates an autonomy gap where an agent could self-enroll into conversations or disengage unexpectedly, increasing the risk of unintended participation, information exposure, or disruption of orchestrated multi-agent workflows.
Content: No source excerpt is available for this finding.