Security audit
LGD Law & Ethic（法律 AI 伦理守门）
Security checks for vulnerabilities and agentic risk
Overview
This skill performs a local self-assessment, but its PASS result should not be treated as legal compliance certification.
Use this as an advisory checklist with human review. Keyword matches and self-reported answers do not establish valid credentials, reliable citations, or legal compliance, and its effectiveness is unvalidated. Use the implemented scripts/law_ethic.py options: --rubric, --system, --answers, and --json. Independent inspection addressed the scanner's missing-reference and image exclusions and found no harmful behavior.
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