Back to skill

Security audit

Response Timing

Security checks for vulnerabilities and agentic risk

Overview

This is a low-risk instruction-only skill that helps an agent decide when to reply in group chat, with a small caution around broad emergency-response guidance.

This skill appears safe to install as an instruction-only response-timing aid. In sensitive or crowded group chats, consider tightening the emergency and follow-up rules so the agent defaults to silence unless directly mentioned or clearly assigned.

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
  • 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
Category
Not specified by scanner
Confidence
90% confidence
Finding

The guidance treats 'obviously a task you are handling' as sufficient to reply, even without a direct mention or explicit re-engagement. In group chat, that heuristic is ambiguous and can cause unsolicited activation, message hijacking, or accidental disclosure into conversations not clearly addressed to this agent.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The emergency exception allows the agent to proactively respond without being addressed whenever a message contains urgency cues like '紧急' or 'ASAP' and seems helpable. This broad unsolicited-activation rule can be abused by users to trigger the agent in unrelated discussions, creating interruption, authority confusion, and possible leakage or unsafe action in sensitive contexts.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.