ClawHub
Back to skill

Security audit

Advisory Committee

Security checks for vulnerabilities and agentic risk

Overview

This is a simple advisory workflow skill with minor quality concerns but no evidence of malicious behavior.

Install only if you want a broad advisory framework for strategy decisions. Review recommendations yourself before acting, and avoid recording sensitive proposal details in the decision log unless you intentionally want them saved in a visible place.

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
  • 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 (3)

Hidden Instructions

High
Category
Prompt Injection
Confidence
88% confidence
Finding

The file begins with an unexpected hidden or non-printing character before the frontmatter. Hidden characters can interfere with parsers, conceal instructions, or create discrepancies between what reviewers see and what the system interprets, which weakens trust in the skill definition even when the visible content appears benign.

Content

Scanner excerpt · SKILL.md (reported line 1)May include surrounding context.

md
---
name: advisory-committee
description: "4-agent AI advisory committee for strategic decisions. Runs CS, Marketing, Finance, and Tech agents in sequence to analyze proposals and deliver a majority-vote recommendation. Trigger: when you need help deciding strategy, prioritization, or whether to build something new."
---

Hidden Instructions

High
Category
Prompt Injection
Confidence
60% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · references/committee-roles.md (reported line 1)May include surrounding context.

md
# Agent Role Definitions

## 🎧 CS Agent "User Champion"

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill's trigger is very broad: it applies whenever a user needs help deciding strategy, prioritization, or whether to build something new. This can cause the skill to activate for a wide range of common decision-support requests, potentially overriding more specific skills or unintentionally steering sensitive business decisions through a fixed multi-agent workflow.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.