Back to skill

Security audit

latticework

Security checks for vulnerabilities and agentic risk

Overview

This skill provides a structured thinking method without granting authority to execute business decisions or access sensitive systems.

Use it as a thinking aid and independently review consequential recommendations. Avoid recording confidential details in its reusable failure-pattern table.

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

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 109)May include surrounding context.

md
## Red Flags

- Only one discipline's vocabulary used throughout
- Stakeholder objections dismissed without checking if they represent another model's prediction
- Decision called "rigorous" because the single model was applied thoroughly
- Diverging data forced into the primary model instead of triggering a model-check
- The analysis cannot name its own blind spots