ClawHub
Back to skill

Security audit

FTM Copilot

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent IBM FTM knowledge skill with no hidden execution, persistence, or data access behavior, though its trigger wording is broader than ideal.

Install this if you want IBM FTM-specific assistance. Be aware that it may activate for nearby enterprise integration topics such as ACE, DB2, MQ, SWIFT, or ISO 20022 even when the task is not specifically about FTM, so users should keep prompts explicit when they want general help instead.

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

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The trigger criteria are overly broad and include an explicit instruction to invoke the skill 'when in doubt,' which can cause unnecessary or unintended activation outside narrowly relevant IBM FTM contexts. This increases the chance that the skill overrides more appropriate routing, injects domain-specific assumptions into unrelated tasks, and expands the attack surface for prompt-manipulation or confused-deputy behavior.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.