Back to skill

Security audit

Norman Mailer Skill

Security checks for vulnerabilities and agentic risk

Overview

This is a text-only writing and ideas skill with no code, persistence, or data access; its main caution is provocative political-risk language.

Install only if you want a provocative literary and political-analysis voice. Treat its advice about direct political or on-the-ground action as rhetorical unless you independently ensure any activity is lawful, nonviolent, consensual, and personally safe.

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
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill explicitly frames political action as physically entering a 'political theater' and valorizes risk-taking, boldness, and unpredictable consequences without any safety, legality, or de-escalation guidance. In an agent setting, this can normalize hazardous real-world conduct and encourage users toward reckless participation in political or confrontational situations.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The example answer endorses 'bringing the body to the scene' and 'going into the现场' as the most powerful form of action, while measuring value by willingness to go deep and assume risk. Because it lacks warnings about personal safety, legality, conflict avoidance, or professional safeguards, it may push users toward unsafe on-the-ground engagement in volatile political or social environments.

Content

No source excerpt is available for this finding.