ClawHub
Back to skill

Security audit

heursistic

Security checks for vulnerabilities and agentic risk

Overview

This skill is a persuasion aid with no executable code, but it broadly encourages manipulative influence tactics without clear ethical limits.

Review carefully before installing. This skill has no technical malware indicators, but it may cause an agent to apply aggressive persuasion tactics too broadly. Use it only for truthful, consent-respecting marketing or sales writing, and avoid deceptive, coercive, regulated, political, interpersonal, or vulnerable-audience scenarios.

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 (2)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The skill strongly promotes psychological influence and repeatedly describes users as more 'vulnerable to influence,' calls understanding bias 'THE WEAPON,' and instructs the system to use these tactics whenever trying to get someone to act, but provides no ethical guardrails or harm warnings. In context, this omission is dangerous because it normalizes manipulative behavior and could enable scams, coercive sales, social engineering, or exploitation of emotional and cognitive biases without consent.

Content

No source excerpt is available for this finding.

Vague Triggers

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill is configured to trigger on essentially any request involving persuasion or getting a human to take action, which is far broader than a narrowly scoped writing aid. That broad activation increases the chance the agent will apply manipulative tactics in inappropriate contexts, including sensitive interpersonal, political, financial, or deceptive scenarios, especially because the content explicitly frames psychological leverage as something to be used 'every time, without exception.'

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.