ClawHub
Back to skill

Security audit

LLM Council

Security checks for vulnerabilities and agentic risk

Overview

This skill is a transparent multi-agent deliberation workflow, with one minor activation-scope concern but no evidence of hidden, destructive, persistent, or credential-seeking behavior.

Before installing, be aware that this skill can launch several subagents and may use search or code inspection when available, so it can add cost and latency. Use it when you explicitly want multiple perspectives or a serious decision review; for simple questions, avoid triggering it accidentally with broad phrasing.

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
97% confidence
Finding

The trigger list includes very broad phrases such as "should I," "am I right about," and "devil's advocate," which commonly appear in ordinary conversation outside the intended multi-agent deliberation use case. This can cause the skill to activate unexpectedly, increasing unnecessary delegation, cost, latency, and the chance that user requests are reframed or over-processed in ways the user did not intend.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.